Anda di halaman 1dari 102



PAGES 259–336
Dyed Purple Ethiopian Opal
Garnet Composition from Gem Properties
Symmetry Boundaries for Round Brilliants
Odontolite in Antique Jewelry

Safety of Irradiated Blue Topaz

NO. 4


Winter 2011 Volume 47, No. 4

259 Great Expectations
Jan Iverson

260 Dyed Purple Hydrophane Opal
Carat Points
Nathan Renfro and Shane F. McClure
Evidence indicates that opal with a vivid purple bodycolor, reportedly from a
new deposit in Mexico, is actually dyed hydrophane from Ethiopia.
272 Determining Garnet Composition from Magnetic Susceptibility
And Other Properties
Donald B. Hoover
Garnet compositions derived from measurements of physical properties
pg. 262
correspond closely with results obtained using chemical data.
286 GIA’s Symmetry Grading Boundaries for Round Brilliant Cut Diamonds
Ron H. Geurts, Ilene M. Reinitz, Troy Blodgett, and Al M. Gilbertson
GIA’s boundary limits for 10 symmetry parameters, measured by optical
scanners, will improve consistency in the symmetry grading of round brilliants.


296 A Historic Turquoise Jewelry Set Containing Fossilized Dentine
(Odontolite) and Glass
Michael S. Krzemnicki, Franz Herzog, and Wei Zhou
Investigation of six antique brooches identifed most of their “turquoise”
cabochons as odontolite.
pg. 287
302 The Radioactive Decay Pattern of Blue Topaz Treated by Neutron Irradiation
Jian Zhang, Taijin Lu, Manjun Wang, and Hua Chen
Some “London Blue” topaz contains radioactive trace impurities that may require
several years to reach a safe level.
308 Lab Notes
Fancy Vivid purple diamond • Strongly purple-colored black diamond • HPHT-treated diamond
with the fluorescence pattern of an HPHT-grown synthetic • Type IIb diamond with long phospho-
rescence • Clarity-enhanced opal with artificial matrix • Ethiopian black opal • Coated bead-cul-
tured freshwater pearls • Tenebrescent zircon
316 Gem News International
Chondrodite from Tanzania • Blue dolomite from Colombia • Fluorite from Namibia • Common opal
from Western Australia • A bicolor, bi-pattern hydrophane opal • Chatoyant quartz with cinnabar
inclusions • Quartz with acicular emerald inclusions • Quartz with izoklakeite inclusions • Ruby and pg. 312
sapphire mining in Pakistan • Color-change sphene from Pakistan/Afghanistan • Cobalt blue–
colored spinel from Vietnam • Trapiche spinel from Mogok, Myanmar • Non-nacreous “cat’s-eye
pearls” • Large synthetic quartz • Sugar-acid treatment of opal from Wollo, Ethiopia • Conference
336 Book Reviews/Gemological Abstracts Online Listing

EDITORIAL Editor-in-Chief Editor and Technical Specialist Associate Editor

STAFF Jan Iverson Brendan M. Laurs Stuart D. Overlin
Senior Manager, Editor, Gemological Abstracts Editors, Lab Notes
Communications Brendan M. Laurs Thomas M. Moses
Amanda Luke Shane F. McClure
Editor, Gem News Contributing Editor Editors, Book Reviews
International James E. Shigley Susan B. Johnson
Brendan M. Laurs Jana E. Miyahira-Smith
Circulation Coordinator Editor-in-Chief Emeritus
Martha Rivera Alice S. Keller
(760) 603-4000, ext. 7142

PRODUCTION Art Director Image Specialist G&G Online:

STAFF Nanette Newbry Kevin Schumacher
Studio 2055

EDITORIAL Ahmadjan Abduriyim Emmanuel Fritsch Robert E. Kane Kenneth Scarratt

Tokyo, Japan Nantes, France Helena, Montana Bangkok, Thailand
Shigeru Akamatsu Jaroslav Hyršl Lore Kiefert James E. Shigley
Tokyo, Japan Prague, Czech Republic Lucerne, Switzerland Carlsbad, California
Edward W. Boehm A. J. A. (Bram) Janse Michael S. Krzemnicki Christopher P. Smith
Chattanooga, Tennessee Perth, Australia Basel, Switzerland New York, New York
James E. Butler E. Alan Jobbins Thomas M. Moses Wuyi Wang
Washington, DC Caterham, UK New York, New York New York, New York
Alan T. Collins Mary L. Johnson Mark Newton Christopher M. Welbourn
London, UK San Diego, California Coventry, UK Reading, UK
John L. Emmett Anthony R. Kampf George R. Rossman
Brush Prairie, Washington Los Angeles, California Pasadena, California

SUBSCRIPTIONS Copies of the current issue may be purchased for $29.95 plus shipping. Online subscriptions are $74.95 for one year (4 issues).
Combination print + online subscriptions are $139.95 in the U.S. and $160 elsewhere for one year. Canadian subscribers should
add GST. Discounts are available for group subscriptions, renewals, GIA alumni, and current GIA students. For institutional
rates, contact the Associate Editor. Subscriptions include G&G’s monthly gemological e-newsletter, the G&G eBrief.
To purchase subscriptions and single issues (print or PDF), visit or contact the Circulation Coordinator.
PDF versions of individual articles and sections from Spring 1981 forward can be purchased at for $12
each. Visit for free online access to the 1934–2010 subject and author index and all 1934–1980 issues.
Gems & Gemology’s five-year impact factor (for 2005–2009) is 1.737, according to the 2010 Thomson Reuters Journal
DATABASE Citation Reports (issued June 2011). Gems & Gemology is abstracted in Thomson Reuters products (Current Contents:
COVERAGE Physical, Chemical & Earth Sciences and Science Citation Index—Expanded, including the Web of Knowledge) and other
databases. For a complete list of sources abstracting G&G, go to
MANUSCRIPT Gems & Gemology welcomes the submission of articles on all aspects of the field. Please see the Guidelines for Authors at
SUBMISSIONS or contact the Editor. Letters on articles published in Gems & Gemology are also welcome.
Abstracting is permitted with credit to the source. Libraries are permitted to photocopy beyond the limits of U.S. copyright law
COPYRIGHT for private use of patrons. Instructors are permitted to photocopy isolated articles for noncommercial classroom use without
AND REPRINT fee. Copying of the photographs by any means other than traditional photocopying techniques (Xerox, etc.) is prohibited with-
out the express permission of the photographer (where listed) or author of the article in which the photo appears (where no pho-
PERMISSIONS tographer is listed). For other copying, reprint, or republication permission, please contact the Editor.
Gems & Gemology is published quarterly by the Gemological Institute of America, a nonprofit educational organiza-
tion for the gem and jewelry industry.
Postmaster: Return undeliverable copies of Gems & Gemology to GIA, The Robert Mouawad Campus, 5345 Armada Drive,
Carlsbad, CA 92008.
Our Canadian goods and service registration number is 126142892RT.
Any opinions expressed in signed articles are understood to be opinions of the authors and not of the publisher.

ABOUT The recent appearance of purple play-of-color opal, reportedly from a new deposit in Mexico, has heightened interest in
THE COVER the origin and nature of this material. In this issue, Nathan Renfro and Shane McClure offer gemological and spectro-
scopic evidence that this opal is actually dyed hydrophane from Ethiopia’s Wollo Province. The fine untreated Wollo
opals on the cover consist of a 35.32 ct oval cabochon in the center courtesy of William Larson (,
Fallbrook, California) surrounded by two pieces of rough (8.16 and 12.78 g) and four polished stones (6.95–32.55 ct)
that are courtesy of David Artinian (Clear Cut Inc., Poway, California). Photos by Robert Weldon.
Color separations for Gems & Gemology are by Pacific Plus, Carlsbad, California.
Printing is by Allen Press, Lawrence, Kansas.

GIA World Headquarters The Robert Mouawad Campus 5345 Armada Drive Carlsbad, CA 92008 USA

© 2011 Gemological Institute of America All rights reserved. ISSN 0016-626X


e all have great expectations for 2012. I’ve gone a step further
with my New Year’s resolutions by creating a “goals” book.
Not only do I write down my goals, but I also add details such
as how and when I will achieve them, and include pictures to
make it more visual.

So, what new things can you expect from Gems & Gemology in the year
ahead? Our goal is to create an enhanced digital journal, one that offers a
more interactive experience for our online audience, in support of what we
already do. Digital is changing the way information is consumed, and it pro-
vides a unique opportunity to reinvent the way we engage with you, our
readers. Digital is beyond relevant: It is the future.

But first we finish 2011. Our final issue of the year includes a report by GIA
researchers on a new purple opal that’s been hitting the market. Although it
was reportedly natural material from Mexico, gemological investigation identified it as dyed opal
from Ethiopia. Because this hydro-
phane opal can be very absorbent,
Our goal is to create an enhanced digital journal, one we can expect to see many other
that offers a more interactive experience for our online treatments applied to it.

audience, in support of what we already do. Another piece by GIA researchers

examines symmetry parameters in
diamond grading. Thanks to improve-
ments in measuring round brilliants with optical scanners, GIA can now evaluate symmetry more
consistently. This measurement-based procedure will complement the visual assessment of symmetry,
which is one of the components of GIA’s diamond cut grade.

We also feature a set of six antique brooches, apparently from the early- to mid-19th century, set
with light blue cabochons that were once thought to be turquoise. Analysis revealed that most of
the supposed turquoise cabochons are actually fossilized dentine, also known as odontolite.

In this issue, we also highlight neutron-irradiated “London Blue” topaz. Some trace impurities in
this topaz become radioactive after neutron irradiation. The authors found
that irradiated samples from China contained up to four of these radioac-
tive impurities. Most of the samples were safe to handle 95 days after irra-
diation, but others will require several years to reach a safe level.

I suspect that many of us, when we were younger, thought we’d be flying
around in space pods by now instead of driving cars. Yet when the future
arrives, we find that technology never changes as fast as our minds imag-
ined it! As we progress in our digital offerings, we are going to keep it sim-
ple by delivering relevant content in innovative ways.

In 2012, I hope you achieve your goals and meet your expectations.


Jan Iverson | Editor-in-Chief |


Nathan Renfro and Shane F. McClure

Opals with an unusual purple bodycolor and strong play-of-color have recently appeared in the
market. Reportedly from a new deposit in Mexico, they have a vivid bodycolor unlike that of any
natural play-of-color opal seen in the trade so far. This alone was enough to raise suspicion, and
gemological and spectroscopic evidence indicates that the purple coloration is artificial. A com-
parison of this purple opal with numerous samples from Ethiopia’s Wollo Province strongly sug-
gests that it is actually dyed hydrophane opal from those deposits. Several previously undocu-
mented characteristics of Wollo opals are described, including zeolite mineral inclusions.

ot often is a significant new deposit discov- has an important implication: Any gem material
ered of a well-known gem material. The that absorbs liquids so easily has the potential to be
1994 discovery of play-of-color opal in treated by methods such as dyeing or impregnation.
Ethiopia’s Shewa Province sparked the industry’s Recognizing this possibility, we performed several
attention. Unfortunately, much of this opal turned experiments to determine the effect of such treat-
out to be unstable, and spontaneous fracturing ren- ments on this opal, before they appeared in the mar-
dered it largely unusable for jewelry (Johnson et al., ket. These experiments—which were surprisingly
1996). In 2008, another large Ethiopian deposit was successful—led us to believe that it would only be a
found in the province of Wollo. While similar in matter of time until we encountered such treated
appearance to some of the Shewa material, this opal material in the trade.
appears to be much more stable than its predecessor Indeed, in October 2011, we were presented with
(Rondeau et al., 2010). The 2011 Tucson show saw several samples of hydrophane opal that had a bright
an abundant influx of beautiful and relatively inex- purple bodycolor (not known to occur naturally in
pensive opal from the Wollo deposit. play-of-color opal; e.g., figure 1), and our suspicions
The new Ethiopian material displayed a property were immediately raised (Renfro and McClure, 2011).
not often seen in opal. Much of the opal is hydro- In addition, these opals were said to be from a new
phane, meaning it is very porous and easily absorbs source in Mexico, but everything about them except
water (or other liquids), often turning translucent or the color reminded us of Ethiopian opal from Wollo.
transparent in the process. The degree to which Purple opal has been reported from several locali-
these stones show this property varies, but some ties, including Mexico. However, all of the material
absorb water so readily that the tiny bubbles escap- examined to date was opaque (or at best translucent)
ing from their surface give the impression of efferves- and did not possess play-of-color. The purple in these
cence. This property, while interesting to watch, also common opals has been attributed to inclusions of
fluorite (Fritsch et al., 2002).
See end of article for About the Authors and Acknowledgments.
While the color of the new play-of-color samples
GEMS & GEMOLOGY, Vol. 47, No. 4, pp. 260–270, was said to be natural, the authenticity of any gem material can only be proven through scientific analy-
© 2011 Gemological Institute of America sis and observation. Our goal in this study is to


answer two fundamental questions: Is this purple
opal naturally colored, and is it actually from a new
deposit in Mexico? This article also reports the
results of an experiment on the dyeing of hydro-
phane opal from Wollo Province.


Nine purple opal cabochons ranging from 4.13 to
15.25 ct, and nine rough opal samples between 0.66
and 1.96 g, were submitted for testing at GIA’s
Carlsbad laboratory. Seven of the rough opals had a
purple bodycolor of varying intensity, but one had a
distinct green-blue bodycolor (0.76 g) and another
was light blue (0.93 g). Client permission was
obtained for potentially destructive testing on these
stones, as hydrophane opal will occasionally crack
when soaked in liquid (author NR’s personal experi-
ence). To expand our sample base, we subsequently
borrowed 22 rough opals (0.57–3.96 g) from the same
source that submitted the purple cabochons. These
ranged from light to very dark purple, except for
three samples showing no visible purple color. Two
of the 22 samples also showed an amber color zone.
We also examined approximately 2 kg of rough natu-
ral hydrophane opal from Wollo (loaned from two
reputable sources who obtained them in Ethiopia)
for comparison with the purple opal.
Figure 1. The purple bodycolor of this new opal, report-
We performed standard gemological characteriza-
edly from Mexico, raised concerns about the origin of
tion of all cut samples (both dyed and natural) with a
the material. Shown here are three rough samples rang-
Duplex II refractometer, a desk-model spectroscope, ing from 1.28 to 3.09 g and five cabochons weighing
a long- and short-wave 4 watt UV lamp, and a gemo- 5.31–9.32 ct. Photo by Robert Weldon.
logical microscope. Inclusions were identified with a
Renishaw InVia Raman microscope using a 514 nm
argon-ion laser at a resolution of 1 cm−1.
Visible spectroscopy measurements were made visually verify that the spectra collected were indeed
by soaking one purple opal cabochon and the green- responsible for the samples’ coloration. Because of
blue piece of rough opal in acetone (with the client’s the very low absorption values of the solutions, they
permission) for 54 hours, and then placing the ace- needed to be concentrated to visually resolve color.
tone in standard 1 cm glass cuvettes for analysis Since we could not physically concentrate our solu-
with a Perkin Elmer Lambda 950 spectrometer. We tions and still have enough volume to fill the
used a data interval and slit width of 1 nm, and base- cuvettes, this was done artificially by multiplying
line correction was accomplished using a 1 cm glass the absorption values by a factor of 15. This method
cuvette filled with pure acetone. Spectroscopy was is equivalent to a solute (dye) concentration 15 times
not performed directly on the opals themselves, in greater than that in the original solutions due to the
order to eliminate interference from light scatter and linearity of absorption. The absorption values of the
the intrinsic opal spectrum that would prevent us artificially “concentrated” acetone solutions were
from making baseline-corrected measurements of converted to transmission spectra using GRAMS
any dye present in the samples. Such measurements spectroscopy software by Thermo; CIE L∗a∗b∗ color
were necessary for color analysis, as described space coordinates were calculated using the GRAMS
below. color analysis application. These coordinates were
From the absorption spectrum of the acetone imported into Adobe Photoshop to produce color
solutions, we calculated color space coordinates to samples.


Figure 2. The purple opal (left, image width 7.8 mm) displayed a “digit” pattern of play-of-color and
a cellular pattern of potch that had a greenish cast, much like natural Wollo opal (right, image
width 4.3 mm). Photomicrographs by S. F. McClure.

Chemical analysis of 15 samples (seven untreated absorbed. After removal from the solution, the sam-
white Wollo opals and eight purple samples) was ples were dried under a tensor lamp for several hours
performed using a Thermo X Series II ICP-MS with a until the acetone had completely evaporated.
New Wave Research UP-213 laser ablation sampling
system and a frequency-quintupled Nd:YAG laser RESULTS AND DISCUSSION
(213 nm wavelength) with a 4 ns pulse width. We Gemological Properties. All the opals provided by our
used 55-µm-diameter ablation spots, a fluence of client were clearly hydrophane, as they tended to feel
around 10 J/cm2, and a 7 Hz repetition rate. Quali- sticky when handled, a result of the opal trying to
tative chemical analysis of two rough samples (one draw moisture from the skin. The spot RI measure-
treated and one untreated) showing black surface ments of the nine purple opal cabochons ranged from
material was also performed with a Thermo ARL 1.37 to 1.41. The SG was between 1.70 and 1.77, as
Quant-X EDXRF system in a vacuum, utilizing no measured hydrostatically before allowing the stones
filter at 4 kV and 1.98 mA, and a cellulose filter at 8 to completely soak full of water. All samples showed
kV and 1.98 mA. a very weak blue reaction to long- and short-wave
As noted above, given the hydrophane character UV radiation. A broadband absorption was seen in
of Wollo opal, it should be amenable to dyeing. To the desk-model spectroscope from ~550 to 600 nm.
test this, eight rough (0.17–1.42 g) and three cabo- Magnification revealed octahedral to irregularly
chons (1.65–3.55 ct) of hydrophane opal from the shaped dark crystals of pyrite, tube-like inclusions
personal collection of author NR were immersed in that resembled fossilized plant matter, and cellular
variously colored solutions prepared from Sharpie play-of-color referred to as a “digit pattern” because
permanent markers and acetone. The samples were of its resemblance to fingers (figure 2; Rondeau et al.,
soaked for anywhere from several minutes to several 2010). The pattern consists of relatively large rounded
hours, depending on how rapidly the solution was cells separated by a thin network of potch (common

Figure 3. In many of the

purple opals, color con-
centrations were evident
in surface pits (left,
image width 2.2 mm),
and on unpolished areas
(right, image width 2.4
mm). Photomicrographs
by S. F. McClure.


ful with rough material. With opal, water serves as
an adequate immersion liquid. One of the rough
stones in the initial group had a light blue bodycolor
and did not appear to be treated. We were quite sur-
prised when immersion revealed a blue zone along
the surface of the stone (figure 5). This type of sur-
face-conformal coloration is indicative of color treat-
ment in many gem materials, including beryllium-
and titanium-diffused corundum, smoke-treated
opal, and dyed agate.

Spectroscopy of the Dye. No color was observed in

the acetone after soaking the purple sample for up

Figure 4. Surface-reaching inclusions in the purple

opals sometimes had a purple color. Those shown here NEED TO KNOW
probably consist of fossilized plant matter. Photo- • In late 2011, purple opal showing strong play-of-
micrograph by S. F. McClure; image width 3.1 mm. color appeared in the market, reportedly from a
new deposit in Mexico.
• The presence of a dye was indicated by soaking
opal without play-of-color) that has a slightly green- the opal in acetone for an extended period (54
ish appearance. Also seen in some samples were sub- hours), followed by spectroscopic processing of
tle purple color concentrations around pits, scratches, the solution that yielded a purple color consis-
surface-reaching fractures, and sometimes on unpol- tent with the opal’s bodycolor.
ished surfaces (figure 3). Some surface-reaching inclu- • Physical, chemical, and microscopic properties
sions were also purple (figure 4). However, not all of of the purple opal overlap those of hydrophane
the samples showed color concentrations. opal from Ethiopia’s Wollo Province, except for
Gem materials are often immersed in a liquid of the purple color.
similar refractive index to see subtle internal fea- • Dye experiments on Wollo hydrophane opal
tures such as color zoning. This is particularly help- produced vivid bodycolors.
• Multiple lines of evidence indicate that the pur-
ple opal is actually dyed Wollo hydrophane.
Figure 5. Immersion of this light blue rough opal in
water showed a surface-conformal color layer, which
is indicative of treatment. Photomicrograph by
N. Renfro; image width 9.5 mm.
to 16 hours. After 54 hours, the acetone solution
appeared very light purple. Spectroscopy of this
sample revealed a broad asymmetrical feature with
an apparent maximum located at about 594 nm and
a shoulder close to 557 nm (figure 6, top). This fea-
ture was consistent with the broadband absorption
observed in the desk-model spectroscope.
The CIE L∗a∗b∗ coordinates calculated using
GRAMS and Adobe Photoshop software yielded a
purple color (figure 6, center), consistent with the
bodycolor of the opal. This proved that the coloring
agent of the purple opal can be partially removed
with acetone, and since we know of no naturally
occurring coloring agent showing this behavior, we
concluded that the opals are colored by an artificially
introduced dye.


To better understand the absorption spectrum of
0.06 557
the purple dye, we used the peak fitting application

0.05 of GRAMS to resolve the individual features from

0.04 the asymmetric absorption (figure 6, bottom). The
full width at half maximum (FWHM) of the 557 nm
feature was 72 nm, and the FWHM of the 594 nm
0.02 feature was 36 nm. Processing of these component
spectra showed that the 557 nm feature was respon-
sible for a purplish pink component, while the 594
400 500 600 700
nm band contributed a blue component. The combi-
nation of these purplish pink and blue features is
responsible for the purple color.
This dye extraction procedure was also applied to
TRANSMISSION SPECTRUM the green-blue piece of rough (figure 7, inset). After
the opal had soaked in acetone for several hours, we
measured the solution’s visible absorption. A broad

band was recorded at 627 nm (figure 7); the calculat-
ed color was again consistent with the opal’s body-


20 1.2
557 627

400 500 600 700



400 500 600 700

594 Figure 7. A green-blue opal was also soaked in ace-

Red component tone, and spectroscopy of the solution revealed a
Blue component
broad absorption feature centered at 627 nm. The cor-
400 500 600 700 responding color space coordinates yielded a color
sample consistent with the bodycolor of the sample.
Inset photo by Robison McMurtry.

Figure 6. After a purple opal was soaked in acetone, Comparison of Dyed Purple Opal to Wollo Hydro-
the solution’s visible-range spectrum revealed a broad phane. We compared the physical appearance, gemo-
asymmetrical absorption feature (top). The absorption logical properties, and other analytical results for the
from the artificially “concentrated” dye (center, trans- dyed material with opals from Wollo Province to
mission spectrum) generated a CIE L*a*b* color
assess the original source of the treated opal and help
space coordinate consistent with the opal’s purple
investigate claims of Mexican origin.
bodycolor. Two color components of the purple dye
were resolved from the transmission spectrum (bot-
tom). The 557 nm feature produced the purplish pink Gemological Properties. The RI and SG values of the
component, and the 594 nm feature contributed the dyed opal were virtually identical to those of Wollo
blue one. opal. This is notable because both properties are par-


Figure 8. Both the Wollo (left) and dyed purple (right) opals showed irregular to octahedral black-appearing crystals
of pyrite. Photomicrographs by N. Renfro (left, image width 0.8 mm) and S. F. McClure (right, image width 1.7 mm).

ticularly low for play-of-color opal (Webster, 1996). purple opal. A similar pattern has been reported in
Ultraviolet fluorescence was also very similar some opal from Virgin Valley, Nevada, but it was
between the two (see also Rondeau et al., 2010). smaller and had a slightly different appearance
Most opal is porous to a minor degree, but it is (Gübelin and Koivula, 2005).
quite unusual for it to be so porous as to qualify as
hydrophane. Both Wollo opal and the purple opal dis- Inclusions. Microscopic characteristics are essential
play this property—sometimes it is so prominent to any comparison of gem materials. All of the fea-
that the transparency can be seen to improve as it tures described in this section were seen in both the
soaks up water. Mexican hydrophane opal is known, untreated Wollo samples and the purple opals.
but to our knowledge it is opaque and light pink or Among those reported previously in Ethiopian opal
brown (with or without play-of-color). are small black octahedral crystals that have been
The structure of the play-of-color is also notewor- suggested to be pyrite (Johnson et al., 1996; Rondeau
thy. Wollo opal sometimes displays an unusual et al., 2010; figure 8) and irregular tubular inclusions
“digit pattern” to its play-of-color that many consid- with a cellular structure that are probably fossilized
er unique to Ethiopian material (Rondeau et al., plant matter (Rondeau et al., 2011; figure 9). Though
2010). We observed this same pattern in some of the neither can be considered unique to Ethiopia, pyrite

Figure 9. Irregular tube-like inclusions that appear to be fossilized plant matter occurred in both the Wollo opal (left,
image width 4.4 mm) and the dyed opal (right, image width 5.0 mm) samples. Photomicrographs by S. F. McClure.


Figure 10. Pseudo-cubic crystals
of a zeolite mineral (possibly
chabazite) were found in both
Wollo opal (left, image width
2.0 mm) and the dyed purple
opal (right, image width 4.3
mm). We believe this is the first
report of this mineral as an
inclusion in opal. Photomicro-
graphs by S. F. McClure.

Figure 11. Surface to subsur-

face circular radiating brown
inclusions were occasionally
visible in both Wollo opal (left,
image width 2.0 mm) and the
dyed purple material (right,
image width 6.5 mm). Photo-
micrographs by S. F. McClure.

octahedra are certainly rare in gem opal. Plant mat- Surface Characteristics. Because we examined
ter is found included in opal from a number of other numerous rough samples of both natural Wollo and
deposits. dyed purple opal, we noted some surface features
We discovered another type of inclusion that to that would not be visible on cut stones.
the best of our knowledge has not been reported in Layers of a dark brown to black opaque material
opal: a zeolite mineral, possibly chabazite. It formed were present on many of the natural Wollo samples.
numerous small, transparent, colorless, euhedral The same material was found on the surface of the
pseudo-cubic crystals (figure 10). They were present rough purple opal we examined—the only difference
only in rough material, at or near the surface, but being that it was distinctly purplish (figure 12). We
always included within the opal, whether the natural were unable to obtain a conclusive Raman spectrum
Wollo or dyed purple material. The crystals appear to from this material, but the spectra did indicate the pres-
have been growing on the matrix before it was ence of amorphous carbon. EDXRF analysis detected
engulfed by the opal. considerable Mn. Previous studies of Ethiopian opal
Also found in the rough opal samples, either at or have identified such material as a manganese oxide
just below the surface, was a flat round brown material (Johnson et al., 1996; Rondeau et al., 2010).
with a radial structure (figure 11). These were typically The rough Ethiopian opal we examined some-
seen along fracture surfaces. We were unable to match times had a brownish pink to pink opaque material
their Raman spectra to anything in our database. on the surface that was very friable and had a matte

Figure 12. Black to brownish

black surface coatings were pres-
ent on rough pieces of both the
Wollo (left) and purple (right)
opals, the only difference being
the purplish cast of the coating on
the dyed material. Photomicro-
graphs by S. F. McClure; image
width 6.7 mm.


Figure 13. Some Wollo opals dis-
played opaque material on the
surface that may be a type of com-
mon opal (left, image width 8.0
mm). Some of the purple samples
showed similar material, except
that it was purplish gray (right,
image width 5.9 mm). Photo-
micrographs by S. F. McClure.

Figure 14. Well-formed quartz

crystals with very short prism
faces were visible in the matrix
of both Wollo opal (left, image
width 2.3 mm) and the dyed
purple material (right, image
width 1.4 mm). Photomicro-
graphs by S. F. McClure.

or dull fracture luster (figure 13, left). Raman spec- (1) transparent light brown hexagonal crystals of
troscopy gave a poor unidentifiable signal, and LA- quartz with very short prism faces (figure 14), some
ICP-MS analysis showed it was composed primarily of which showed dissolution features and contained
of Si. We suspect it is some kind of highly disordered rounded light green and black inclusions (figure 15);
common opal. This same material was found on the (2) prismatic colorless to light yellow crystals of K-
surface of some of the rough purple opal, the only feldspar; and (3) dark green prismatic fractured crys-
difference being the color: It was mostly dark gray tals of aegirine, a pyroxene (figure 16). The quartz
and uneven, with much darker fractures and an and feldspar were common, but we observed only
often purplish cast (figure 13, right). one example of the aegirine in a Wollo opal and one
in a purple sample.
Matrix. Many of the rough samples, both natural One question that arises is how the purple opal
Wollo and dyed purple, still had matrix attached to can be dyed if the matrix of some pieces is still the
them. The matrix was typically beige, though sever- natural beige color. Digging into an area of dark pur-
al pieces of the dyed opal had matrix that was very plish gray matrix with a needle probe showed the
dark purplish gray. In all cases, it was a soft clay-like normal beige color just below the surface (figure 17).
material that contained numerous mineral grains. This indicates that the dye did not penetrate the
These matrix-hosted mineral grains were found in matrix very deeply. Therefore, one possible explana-
both the Wollo and the purple opal, and consisted of: tion for the natural-colored matrix on the purple

Figure 15. Some of the etched quartz

attached to the matrix of both kinds
of opals contained identical inclu-
sions of a translucent green material
and a black mineral (too small to
identify with Raman analysis).
Photomicrographs by S. F. McClure;
image width 1.4 mm (left, Wollo
opal; right, dyed purple opal).


Figure 16. Aegirine crys-
tals were found in the
matrix of one Wollo (left,
image width 2.2 mm)
and one dyed purple opal
(right, image width 1.9
mm). Photomicrographs
by S. F. McClure.

opal is that the top layer of matrix was removed The small number of samples (15 total) analyzed
after the stones were dyed. The softness of the for this study makes it difficult to detect meaningful
matrix would make this easy to do. chemical trends. Overall, the Ca content of all sam-
ples was much higher than the threshold reported by
Chemical Composition. We suspected that chemi- Gaillou et al. (2008): 8,000–10,000 ppmw or higher.
cal analysis would provide important clues to the Additionally, all the samples contained trace amounts
origin of the purple opal. Gaillou et al. (2008) found of Nb. The Ba content of the purple opals tended to be
that opal from the initial Ethiopian deposits at lower (80–155 ppmw) but reached 475 ppmw. The Ba
Shewa was easily distinguished by its high Ca con- content of the untreated Wollo opal was higher over-
tent (>1000 ppmw), combined with the presence of all (175–285 ppmw), but ranged from 65 to 1400
Nb. That study found high Ca in orange to brown ppmw. There are several possible explanations for
opals only—white Ethiopian opals were not dis- this, one of which is that we do not know the starting
cussed. Two years later, in a report on Wollo opal, color of the dyed opal. It could have been yellowish or
Rondeau et al. (2010) noted a high Ba content (>100 orangy, in which case it would be expected to have a
ppmw) in white opals. This was based on a very lower Ba content, as reported by Rondeau et al. (2010).
small sample base, and the article acknowledged None of the other elements showed any meaningful
that further analyses were necessary. trends. Clearly, further investigation is necessary to
assess the compositional range of these opals.

Figure 17. In some instances, areas of matrix still

attached to the rough purple opal had a dark pur- Dyeing Ethiopian Hydrophane. Vivid bodycolors
plish gray color. When scratched with a probe, the were produced in our dye experiments (figure 18),
matrix under the surface displayed the normal beige and they show how easy it is to artificially color
color. Photomicrograph by S. F. McClure; image Ethiopian hydrophane opal. We also noted dye con-
width 5.9 mm. centrations around fine scratches and pits in the sur-
face, implying that the dye transport mechanism is a
function of surface area. As scratches represent local-
ized zones of high surface area, it is reasonable to
conclude that the transport of a mobile fluid in the
opal is driven by capillarity.
To support this hypothesis, we scratched “GIA”
on one white Ethiopian opal cabochon and immersed
it in our blue dye solution for several seconds. The
dye penetrated the stone much deeper in the area
immediately surrounding the scratched letters (figure
19, left and center). Very fine scratches on the cabo-
chon were also visibly colored by the blue dye. We
then tested the stability of the dyed color by soaking
the stone in pure acetone. After several hours, the
acetone turned slightly blue and the stone became
more uniformly colored and lighter (figure 19, right).


A second dyed blue opal was soaked in water for
more than a week with no observable change in the
color of the opal or the water. The stability of the
dye is therefore largely dependent on the type of sol-
vent to which it is exposed.
Interestingly, the client discovered that soaking
the purple opals in hydrogen peroxide would decol-
orize them. With the client’s permission, we demon-
strated this on a 0.57 g piece of rough purple opal,
soaking it in a 3% solution of hydrogen peroxide for
72 hours. A significant amount of color was lost from
the stone, which had a light purplish gray appearance
after soaking (figure 20). An absorption spectrum col-
lected on the hydrogen peroxide after soaking the
purple opal was featureless. This suggests that the
dye was not necessarily removed from the opal or dis-
solved into the hydrogen peroxide solution; instead,
the dye molecules were chemically altered into a
compound that does not absorb visible light, or
bleached. This technique may be effective in decol-
orizing some dyed opal, but it would be highly depen-
dent on the type of dye used. Figure 18. A variety of vivid colors were easily pro-
duced by dyeing white Wollo opals with solutions
Identification and Origin of Dyed Purple Opal. prepared from permanent marker ink and acetone.
Purple color has never been reported for natural The cabochon in the center was submitted by a client
play-of-color opals. Microscopic examination and weighs 8.77 ct. The rough samples weigh
0.17–1.42 ct; photo by C. D. Mengason.
strongly suggested the presence of a dyeing agent in
the purple sample we examined: color concentra-
tions in fractures and surface pits, patches of purple removed with acetone.
color on the surface, surface-reaching inclusions The purported Mexican origin of the purple opal
that were purple, and certain characteristics on the can be dismissed. Instead, the purple material pre-
surface of the rough such as opaque material with sents all the characteristics of opals from the Wollo
dark purplish gray fractures and layers of a dark deposit in Ethiopia. Both opals are hydrophane,
brown to black material with a purple cast. Proof of sometimes showing a cellular play-of-color separated
dye treatment was that the color could be partially by greenish internal potch. Some contain tiny octa-

Figure 19. This 1.75 ct Wollo opal with “GIA” scratched onto the surface shows how dye transport is
faster in localized regions with a high surface area, such as pits and scratches (left and center). After the
opal was soaked in pure acetone for several hours, the blue dye was homogeneously distributed, elimi-
nating visible dye concentrations (right). Photos by N. Renfro; center image width is 3.8 mm.


Figure 20. This 0.57 g rough
piece of opal is shown before
(left) and after (right) its purple
color was removed by soaking
in hydrogen peroxide for 72
hours. Photos by Robison

hedral inclusions of pyrite and rod- or tube-like the purple samples examined by GIA were dyed
inclusions that are probably fossilized plant matter. opals from Ethiopia. The distinct hydrophane
They have similar RI, SG, and UV fluorescence nature of Ethiopian opal makes it susceptible to
characteristics, as well as surface layers of man- many kinds of treatment—dyeing is just one.
ganese oxide. In addition to these properties, we Smoke treatment of this opal to turn it black was
observed some unusual inclusions in both Wollo recently described (Williams and Williams, 2011),
and dyed purple samples that have not been report- as was sugar treatment (see the Gem News
ed previously, such as transparent pseudo-cubic International entry on pp. 333–334 of this issue).
crystals of a zeolite mineral (probably chabazite) Dyeing this material to more believable colors,
and flat radial brown inclusions. such as the orange of Mexican fire opal, may pose
new identification challenges, and we can expect to
CONCLUSION see more treatments applied to this hydrophane
The evidence presented in this study indicates that opal in the future.


Mr. Renfro ( is staff gemologist, Poway, California) for supplying untreated samples of
and Mr. McClure is director of Identification Services, at Wollo opal. Hussain Rezayee (Rare Gems & Minerals,
GIA’s laboratory in Carlsbad. Beverly Hills, California) provided additional samples.
We thank Evan Caplan for his insights and informa-
ACKNOWLEDGMENTS tion, and Dr. James Shigley for his assistance with the
The authors are grateful to David Artinian (Clear Cut Inc., manuscript.

Fritsch E., Ostrooumov M., Rondeau B., Barreau A., Albertini D., Research, Sept. 26,
Marie A.-M., Lasnier B., Wery J. (2002) Mexican gem opals— from-research/Dyed%20Purple%20Opal%201004.pdf.
Nano and micro structure, origin of colour, and comparison Rondeau B., Fritsch E., Mazzero F., Gauthier J., Cenki-Tok B.,
with other common opals of gemological significance. Bekele E., Gaillou E. (2010) Play-of-color opal from Wegel
Australian Gemmologist, Vol. 21, No. 6, pp. 230–233. Tena, Wollo Province, Ethiopia. G&G, Vol. 46, No. 2, pp.
Gaillou E., Delaunay A., Rondeau B., Bouhnik-le-Coz M., Fritsch 90–105,
E., Cornen G., Monnier C. (2008) The geochemistry of gem Rondeau B., Fritsch E., Bodeur Y., Mazzero F., Cenki T., Bekele E.,
opals as evidence of their origin. Ore Geology Reviews, Vol. 34, Ayalew D., Cenki-Tok B., Gauthier J.-P. (2011) Wollo opals—A
pp. 113–126, powerful source from Ethiopia. InColor, No. 17 (Summer), pp.
Gübelin E., Koivula J. (2005) Photoatlas of Inclusions in 24–35.
Gemstones, Vol. 2. Opinio Publishers, Basel, Switzerland. Webster R. (1996) Gems: Their Sources, Descriptions and Identi-
Johnson M.L., Kammerling R.C., DeGhionno D.G., Koivula J.I. fication, 5th ed. Butterworth-Heinemann Ltd., Oxford, UK.
(1996) Opal from Shewa Province, Ethiopia. G&G, Vol. 32, No. Williams B., Williams C. (2011) Smoke treatment in Wollo opal.
2, pp. 112–120,
Renfro N., McClure S. (2011) A new dyed purple opal. News from [date accessed: Sept. 22, 2011].


Spring–Winter 2010

Spring 2007 Summer 2009

Pink-to-Red Coral: Determining Origin of Color Celebrating 75 Years of Gems & Gemology Now Available
Serenity Coated Colored Diamonds
Trapiche Tourmaline from Zambia
The “Type” Classification System of Diamonds
Spectral Differentiation Between Copper and Iron Online:
Summer 2007 Colorants in Gem Tourmalines
Global Rough Diamond Production since 1870 Andalusite from Brazil All Articles
Durability Testing of Filled Diamonds Peridot from Sardinia, Italy and Issues 1981–2011
Chinese Freshwater Pearl Culture Fall 2009
Yellowish Green Diopside and Tremolite from Characterization of “Green Amber”
Tanzania Crystallographic Analysis of the Get PDF Articles at
Polymer-Impregnated Turquoise Tavernier Blue
Fall 2007 “Fluorescence Cage”: Visual Identification
Electronic (PDF) versions of all articles from
The Transformation of the Cultured Pearl Industry of HPHT-Treated Type I Diamonds
Nail-head Spicule Inclusions in Natural Gemstones Ammolite Update Spring 1981 forward are available as part of
Copper-Bearing Tourmalines from New Deposits Polymer-Filled Aquamarine Gems & Gemology Online.
in Paraíba State, Brazil Yellow-Green Haüyne from Tanzania
Type Ia Diamond with Green-Yellow Color Due to Ni Aquamarine from Masino-Bregaglia Massif, Italy Order Back Issues (Print and PDF)
Winter 2007 at
Winter 2009 (PDF only)
Latest CVD Synthetic Diamonds from Apollo or Call Toll Free 800-421-7250 ext. 7142
Ruby and Sapphire Production and Distribution:
Diamond Inc. or 760-603-4000 ext. 7142
A Quarter Century of Change
Yellow Mn-Rich Tourmaline from Zambia Fax 760-603-4070
Cutting Diffraction Gratings to Improve
Fluorescence Spectra of Colored Diamonds
Dispersion (“Fire”) in Diamonds
An Examination of the Napoleon Diamond Necklace E-mail
Chrysoprase and Prase Opal from Haneti,
Spring 2008 Central Tanzania or write to
Copper-Bearing (Paraíba-type) Demantoid from Val Malenco, Italy Gems & Gemology
Tourmaline from Mozambique
PO Box 9022, Carlsbad, CA
A History of Diamond Treatments Spring 2010
Natural-Color Purple Diamonds from Siberia
92018-9022, USA
Strongly Colored Pink CVD Lab-Grown
Summer 2008 Diamonds Complete volumes of 1992–2011 print
Emeralds from Byrud (Eidsvoll), Norway Color Alterations in CVD Synthetic Diamond back issues (except 2009) are available,
Creating a Model of the Koh-i-Noor Diamond with Heat and UV Exposure
as are limited issues from 1985–1991.
Coated Tanzanite Possible “Sister” Stones of the Hope Diamond
Coloring of Topaz by Coating and Confocal Micro-Raman Spectroscopy 10% discount for GIA Alumni and active
Diffusion Processes Bastnäsite-(Ce) and Parisite-(Ce) from Malawi GIA students.
Fall 2008 Summer 2010
Identification of Melee-Size Synthetic The Wittelsbach-Graff and Hope Diamonds:
Yellow Diamonds Not Cut from the Same Rough
Aquamarine, Maxixe-Type Beryl, and Play-of-Color Opal from Ethiopia
Hydrothermal Synthetic Blue Beryl A New Type of Composite Turquoise
A New Type of Synthetic Fire Opal: Fire Opal from Madagascar
Mexifire X-ray Computed Microtomography
The Color Durability of “Chocolate Pearls” Applied to Pearls
Hibonite: A New Gem Mineral
Winter 2008
Color Grading “D-to-Z” Diamonds at the GIA
Fall 2010
An Era of Sweeping Change in Diamond and
Rubies and Sapphires from Winza, Tanzania
The Wittelsbach Blue Colored Stone Production and Markets
Gem Localities of the 2000s
Spring 2009 Gemstone Enhancement and Its Detection
The French Blue and the Hope: New Data in the 2000s
from the Discovery of a Historical Lead Cast Developments in Gemstone Analysis Techniques
Gray-Blue-Violet Hydrogen-Rich Diamonds and Instrumentation During the 2000s
from the Argyle Mine
Hackmanite/Sodalite from Myanmar and Winter 2010
Afghanistan Synthetic Gem Materials in the 2000s
Pink Color Surrounding Growth Tubes and Yellow Scapolite from Madagascar
Cracks in Tourmalines from Mozambique Pietersite from Namibia and China
Identification of the Endangered Pink-to-Red Update on Mexifire Synthetic Fire Opal
Stylaster Corals by Raman Spectroscopy Gems in a Ciborium from Einsiedeln Abbey

For a complete list of articles from 1981 forward, visit

Donald B. Hoover

Quantitative measurements of magnetic susceptibility combined with RI or SG data can provide

an easy and inexpensive way of inferring garnet composition. At the time this technique was first
applied to faceted garnets (Hoover et al., 2008), a reference set of samples with well-character-
ized compositions was not available. GIA subsequently provided 28 garnets and their chemical
data determined by electron microprobe for a comparison with end-member compositions calcu-
lated from magnetic susceptibility measurements and other properties. The results show that end-
member compositions based on microprobe data have error margins similar to most of those
derived from directly measured properties.

arnet is a diverse gem material showing an measures, in combination with color and spectro-
attractive palette of colors (e.g., figure 1) and scopic data, to infer a garnet composition that is
incorporating a variety of chemical compo- most probable. These compositions were limited to
nents that are responsible for widely ranging values of one or two end members; for garnets in which three
physical properties. The several end-member species or more end members were important, gemologists
may occur as nearly pure compositions or, more com- had no effective recourse.
monly, complex assemblages. The principal species In recent years, with the availability of very
(table 1) are pyrope, almandine, and spessartine (pyral- strong rare-earth magnets, gemologists have started
spite garnets), and grossular, andradite and minor to apply magnetic attraction as a tool for gem identi-
uvarovite (ugrandite garnets). Other end-member fication (see, e.g.,
species, including goldmanite (vanadium rich), knor- Although all materials respond to an applied mag-
ringite (chromium rich), and schorlomite (titanium netic field in some way (box A), it is the transition
rich) also may be present in small amounts, and these elements in garnet that give rise to a measureable
are mainly important for their effect on garnet col- magnetic attraction (reported here as the volume
oration. magnetic susceptibility) if they are present as princi-
In the past, gemologists have been limited in pal components. Recently, Hoover and Williams
their ability to determine garnet composition by (2007) developed a simple, inexpensive apparatus to
only having RI and possibly SG data as quantitative measure volume susceptibility on cut gems (box B).
Hoover et al. (2008) derived garnet composition from
plots of RI versus susceptibility, and followed the
See end of article for About the Author and Acknowledgments.
conventional characterization of Stockton and
GEMS & GEMOLOGY, Vol. 47, No. 4, pp. 272–285, Manson (1985) in defining garnet varieties (pyrope,
© 2011 Gemological Institute of America pyrope-almandine, almandine, almandine-spessar-


Figure 1. This
photo shows the
28 GIA garnet
samples used in
this study
(0.18–9.82 ct).
Photo by Robert

2486 27423 491

26620 12588 2491
13113 13047
11090 25867
11089 234
26767 5821
27352 77
996 4967 13234
11568 27257

tine, etc.). Furthermore, the magnetic susceptibility accuracy of garnet compositions derived from the
technique permitted a quantitative measure of gar- various properties is assessed.
net composition consisting of three end members,
not two, when RI was the only other data available. MATERIALS AND METHODS
Unfortunately, Hoover et al. (2008) did not have gar- Materials. GIA initially supplied data for 539 garnets
net samples of known composition to test how well that included color, carat weight, RI, SG, cell con-
the technique agreed with quantitative chemical stant, and variety, although the data set was incom-
analysis. Using selected samples from the large plete for a number of the stones. The author then
group studied by Manson and Stockton, this article borrowed 28 of the samples for magnetic susceptibili-
compares garnet compositions from GIA’s electron ty measurements (see figure 1 table 2) that were
microprobe data to those inferred from the GIA- selected to cover the full range of compositions and
measured properties that were combined with the RI values. An additional constraint was that each
author’s measured magnetic susceptibilities. The stone be large enough for good susceptibility mea-

TABLE 1. Silicate garnet end-member species and their properties.a

Cell Volume susceptibility
End member Symbol Formula RI SG constant (Å) (×10 −4 SI)

Pyrope Prp Mg3Al2Si3O12 1.714 3.58 11.459 −0.2

Almandine Alm Fe3Al2Si3O12 1.829 4.32 11.528 40.7
Spessartine Sps Mn3Al2Si3O12 1.799 4.20 11.614 47.5
Grossular Grs Ca3Al2Si3O12 1.734 3.59 11.851 −0.2
Andradite Adr Ca3Fe2Si3O12 1.887 3.86 12.048 30.8
Uvarovite Uv Ca3Cr2Si3O12 1.865 3.85 11.996 12.9
Goldmanite Go Ca3V2Si3O12 1.834 3.77 12.070 6.9
Knorringite Kn Mg3Cr2Si3O12 1.875 3.84 11.622 13.7
a See text for sources of data.


TABLE 2. Garnet samples used in this study, their measured and calculated properties, and their compositions.a
Measured Calculated (Locock)
Sample Weight Cell Volume Cell Volume
no.b Variety (ct) RI SG constant susceptibility Composition RI SG constant susceptibility Compositionc
(Å) (×10 −4 SI) (Å) (×10 −4 SI)

77 Rhodolite 2.42 1.752 3.83 11.493 16.0 Prp63Alm22Sps15 1.752 3.83 11.494 13.5 Prp63Alm32Grs3Sps1
79 Color-change 1.88 1.751 3.85 11.583 16.4 Prp63Sps20Alm17 1.752 3.82 11.571 15.9 Prp48Sps23Grs15Alm12Uv1
198 Mint green 2.73 1.735 3.61 11.850 0.4 Grs99Adr1 1.736 3.60 11.850 0 Grs94Prp1Adr1Go1
234 Malaya 2.27 1.765 3.91 11.549 24.6 Prp47Sps32Alm21 1.766 3.93 11.545 23.7 Prp43Sps34Alm18Grs4Adr1
491 Malaya 1.53 1.762 3.90 11.555 24.7 Prp47Sps46Alm7 1.759 3.88 11.558 21.8 Prp45Sps38Alm8Grs7Adr1
996 Almandine- 6.78 1.810 4.22 11.580 44.9 Sps63Alm37 1.811 4.24 11.580 44.4 Sps57Alm40Grs1
2211 Pyrope- 2.14 1.762 3.87 11.530 16.2 Prp52Alm40Grs8 1.763 3.88 11.529 16.5 Prp50Alm39Grs9Adr1Sps1
2486 Rhodolite 2.20 1.762 3.85 11.509 15.7 Prp52Alm36Grs12 1.759 3.86 11.508 15.3 Prp57Alm36Grs4Sps1Adr1
2489 Hessonite 1.25 1.755 3.65 11.889 5.8 Grs83Adr9Alm1 or 1.760 3.64 11.882 5.0 Grs80Adr16Prp1
2491 Demantoid 1.45 1.881 3.84 — 29.7 Adr96Grs4 1.887 3.86 12.048 30.8 Adr96
3429 Pyrope- 3.55 1.784 4.02 11.508 22.9 Alm57Grs33Prp10 1.781 4.00 11.510 23.6 Alm55Prp40Sps2Grs2
4967 Pyrope- 3.18 1.750 3.82 11.534 12.1 Prp66Alm30Grs4 1.751 3.80 11.529 11.8 Prp59Alm28Grs11Adr1
5821 Almandine- 2.52 1.810 4.19 — 46.6 (Outside the 1.804 4.22 11.601 46.3 Sps83Alm16
spessartine ternaries)
11089 Chrome- 1.96 1.744 3.70 11.537 6.2 Grs57Prp27Alm16 1.740 3.70 11.521 6.8 Prp72Alm14Grs7Uv6Adr1Sps1
11090 Chrome- 5.37 1.742 3.72 — 8.0 Prp58Grs22Alm20 1.742 3.71 11.533 6.9 Prp71Alm14Grs7Uv6Adr1Sps1
11568 Pyrope- 3.39 1.807 4.15 — 31.3 Alm77Grs19Prp4 1.804 4.16 11.525 32.4 Alm74Prp19Sps4Grs2
12487 Spessartine 1.41 1.800d 4.23 — 47.0 Sps97Alm3 1.805 4.22 11.580 45.6 Sps72Alm26
12588 Almandine- 1.62 1.812 4.26 — 44.3 Sps53Alm47 1.812 4.25 11.560 43.3 Alm52Sps47Grs1
13047 Spessartine 1.91 1.800d 4.20 — 52.3 (Outside the 1.805 4.22 11.581 45.4 Sps72Alm26
13113 Chrome- 1.08 1.732e 3.74 11.535 8.0 Prp58Grs22Alm20 1.744 3.73 11.530 7.9 Pyp69Alm17Grs7Uv5Adr1Sps1
13234 Demantoid 1.79 1.882 3.87 — 28.8 Adr90Uv7Grs3 1.887 3.86 12.048 30.8 Adr98
25000 Pyrope 1.27 1.744 3.77 11.492 10.9 Prp74Alm25Sps1 1.745 3.77 11.495 10.2 Prp68Alm24Grs3Adr1
25867 Pyrope 0.18 1.730 — — 4.5 Prp79Alm12Grs9 1.733 3.68 11.512 5.3 Prp77Alm12Grs8Adr1Uv1
26620 Almandine 3.06 1.791 4.13 11.534 27.8 Alm64Prp32Sps4 1.795 4.10 11.531 28.8 Alm67Prp24Grs5Sps3
26767 Almandine 0.52 1.793 4.10 11.513 25.3 Alm62Grs35Prp3 1.798 4.13 11.513 30.4 Alm73Prp24
27257 Spessartine 1.42 1.800 d 4.25 — 46.6 Sps97Alm3 1.805 4.22 11.580 45.4 Sps69Alm26
27352 Hessonite 9.82 1.754 3.64 — 4.3 Grs87Adr13 1.757 3.64 11.875 4.8 Grs84Adr12Alm1Sps1
27423 Hessonite 9.10 1.755 3.65 — 3.9 Grs86Adr14 1.752 3.63 11.872 3.4 Grs86Adr12
Data that was not available from GIA is indicated by “–”.
b Numbers in parentheses are the former catalog numbers.
c Note that calculations from the oxide chemistry seldom give end-member compositions that add to 100%.
These RI values are problematic; see text.
e This RI was rechecked by the author and determined to be 1.742.

surements (e.g., >1 ct was preferred). When available, where X is commonly Ca2+, Mn2+, Fe2+, and/or
samples with measured SG and cell constant data Mg2+, and Y is commonly Al3+, Fe3+, Cr3+, and/or
were used; color was not part of the selection process. V3+. Because garnets form an isomorphous series,
the X and Y positions can hold any combination of
Determination of End-Member Compositions. Sili- the respective ions listed; substitutions may also
cate garnets have the general formula X32+Y23+Si3O12, occur for Si.


There are two basic ways to determine the end-
member composition of garnets: calculation from NEED TO KNOW
chemical data and derivation from measured proper-
ties. Chemical data (such as from an electron micro- • Combined with RI or SG data, magnetic suscep-
probe or wet chemical analysis) are typically tibility is one more measureable property that is
useful toward inferring garnet composition.
expressed as wt.% oxides, and there are numerous
(nonequivalent) ways to calculate end-member • Magnetic susceptibility of faceted gemstones can
be measured nondestructively with a relatively
compositions from such data. A commonly used
simple apparatus.
procedure from Rickwood (1968) was slightly modi-
• Susceptibility measurements are plotted against
fied by Manson and Stockton (1981). Rickwood
other properties on modified Winchell diagrams
(1968) discussed the variations that can arise from to derive garnet composition.
the different calculation methods, using a common
• A comparison of garnet compositions derived
metamorphic garnet composition of Prp44Alm42Grs17 from measured properties versus chemical data
as an example, in which the pyrope content can showed a fairly good correlation.
vary by 3.4%, the almandine by 3.4%, and the
grossular by 5.2%, depending on how the calcula-
tions are done. A more recent procedure by Locock
(2008) incorporated advances in the understanding measure of the quality of the analysis. For this
of the crystal chemistry of natural garnets through a paper, the author used the Locock procedure to


Of greatest interest to gemology are the para-
A ll material substances react to the presence
of a magnetic field: They develop an induced
magnetic field in response to the applied field.
magnetic materials and their susceptibilities,
which can have some diagnostic value. By con-
The ratio of the induced field to the applied field trast, diamagnetic susceptibilities have little
is called the volume susceptibility (k) of the sub- diagnostic value. Ferromagnetic materials, when
stance. It is a simple dimensionless ratio. present as inclusions in gems, can give anoma-
Materials react to a magnetic field in three lously high values of magnetic susceptibility for
different ways. Most materials are very weakly the host material.
repelled, or diamagnetic. In this case, k is nega- Physicists have defined several different kinds
tive. A material with a sufficient number of of magnetic susceptibility. Although volume
atoms of the transition elements (Fe, Mn, Cu, Cr, magnetic susceptibility is dimensionless, the
etc.) or the rare-earth elements—depending on numeric value will differ with the system of units
their valence state—may overcome the diamag- being used, changing by a factor of 4π, or 12.57.
netic effects of the other atoms and be attracted Thus the system of units needs to be stated, even
to a magnet. For these materials, k is positive. If for this dimensionless number. This article uses
the value of k is independent of the strength of the International System of Units (SI), which may
the applied field, the material is called paramag- be a possible source for confusion if one is not
netic. Here, k will be positive and of small to familiar with this peculiarity in some electromag-
intermediate magnitude. If k changes with the netic measurements. Another commonly used
strength of the applied magnetic field, the materi- property is mass magnetic susceptibility, also
al is ferromagnetic, and k can be very large. called specific susceptibility. This measure has
Ferromagnetic materials are further divided into dimensions of inverse density (e.g., cubic cen-
true ferromagnetic, ferrimagnetic, antiferromag- timeters per gram), and again one needs to be
netic, and canted antiferromagnetic. These varia- aware of a multiplier of 4π when other units of
tions in behavior are due to interactions between measurement are used. Molar magnetic suscepti-
the electrons in the material and the formation of bility may also be expressed in units of cubic cen-
what are called magnetic domains. timeters per mol, or their equivalent.


13234 Adr

1.880 2491




Figure 2. This plot of RI

vs. magnetic susceptibili- Go

ty (RIMS diagram) com- 1.830


pares measured properties

(indicated by red trian- 1.820

gles) against properties
calculated from end- 1.810 11588

member compositions 996 5821

(green triangles, according 1.800 26767

27257 Sps
to Locock [2008]) for the 26620
28 GIA garnet samples. 12487

The black squares repre- 3429

sent pure end-member
properties (see table 1 for 1.770
key to abbreviations). The 2486
pyralspite and ugrandite 1.760

ternary triangles are 27423

77 79
shown by black lines con- 1.750
necting the corresponding 25000

end members. 1.740

198 11089
Grs End member
1.730 13113
25867 Measured value
1.720 Locock-derived value

0 5 10 15 20 25 30 35 40 45 50 55


obtain the end-member compositions from the GIA where RI = refractive index, S = magnetic suscepti-
oxide chemical data. bility, m = measured, EM = end-member values, and
The second way to determine garnet end-mem- A, B, and C = percentages of end members. With two
ber composition is to use quantitative measured measured properties, one can solve for three possible
properties and solve a series of equations that are end members. With three measured properties, the
based on Vegard’s law, which showed that garnet end members increase to four, and so on. As with
properties are additive functions of the molar propor- compositions based on chemical analyses, the result
tions of end-member compositions (Hutchison, is not unique; there will be several (similar) possibili-
1974). The equations are: ties. Winchell (1958) showed how these equations
can be solved graphically in a rather simple way, and
1. RIm = ARIEM1 + BRIEM2 + CRIEM3 his diagram of RI vs. unit cell dimensions (or cell
2. Sm = ASEM1 + BSEM2 + CSEM3 constant) demonstrated the interrelation between
the pure end members and a particular garnet.
3. A + B + C = 1 However, the cell constant of an unknown garnet is


not easily obtained by the gemologist because X-ray Figure 2 is the modified Winchell diagram plot-
diffraction data is required, so Hoover et al. (2008) ting RI vs. magnetic susceptibility (RIMS). Eight gar-
modified the Winchell diagram so that the composi- net end members of gemological importance (black
tion of an unknown garnet can be determined— squares) are shown on the diagram. The garnet
according to three or four end members—from quan- ternaries pyralspite (pyrope, almandine, and spessar-
titative measurements of properties such as magnet- tine) and ugrandite (uvarovite, grossular, and andra-
ic susceptibility, RI, and SG. Box B describes how dite) are shown as triangles outlined in black that
magnetic susceptibility was measured, both in this connect each of the three corresponding end mem-
study and by Hoover et al. (2008). bers. Other end members shown are goldmanite and


W hile there are several ways to quantitative-

ly measure magnetic susceptibility, vol-
ume susceptibility is routinely measured with a
with the gem’s table. The gem was then slightly
separated from the magnet to obtain a maximum
change in weight (i.e., weight loss for a paramag-
Gouy balance or the similar Evans balance (see netic gem). The procedure is no more complex than mea-
Susceptibility is measured by placing a sample on suring specific gravity—and takes about as much
one arm of a laboratory balance and subjecting it time. The apparatus can be constructed at low cost
to a strong magnetic field gradient. The weight using a surplus microscope.
loss or gain is measured and converted to suscepti-
bility. Unfortunately, the sample must be in the
form of a cylinder. In practice, the sample is often
ground to a powder and placed in a cylindrical
sample holder. This obviously is not practical for
gem materials.
Hoover and Williams (2007) showed that if a
very strong permanent magnet is used, and its Magnet
pole face is smaller than the table (or other flat Sample
facet) of the gem to be measured, then the force of
attraction between the magnet’s pole face and the
facet will be proportional to the gem’s volume
susceptibility. To calculate the gem’s susceptibili-
ty, the magnet only needs to be calibrated with a Digital scale
material of known susceptibility. The apparatus
used by Hoover and Williams (2007), Hoover et al. X-Y stage
(2008), and in this study consisted of a biological
microscope with the optics removed (figure B-1).
In place of the optics, a small iron rod was fixed
vertically to hold a variety of small (1⁄16 in. to 1⁄4 in.
diameter) cylindrical rare-earth magnets. The
important components are the fine focus mecha-
nism (for precise control of the magnet’s vertical Focus adjustment
position) and the moveable X-Y stage that is used
to align the gem’s table with the magnet’s pole
face. A small digital scale was placed on the Figure B-1. This instrument was designed by
microscope stage, and a gem was placed on a Hoover and Williams (2007) for taking magnetic
pedestal in the weighing dish. The magnet’s pole susceptibility measurements, and was used in
face was brought just into contact and parallel the current study. Photo by Bear Williams.


knorringite. (Schorlomite is not shown because end- on synthetic samples by Ringwood (1977), or with
member property values are not available in the the data presented by Nixon and Hornung (1968),
mineralogical literature, and in gem garnets this who first defined knorringite. The McConnell data
component may be present in only very small quan- will continue to be used in this article, with the
tities.) The positions of all the garnets obtained for understanding that knorringite end-member values
this study are also plotted: the green triangles repre- are subject to change.
sent the compositions calculated from microprobe Magnetic susceptibility values are not well
analyses (using the Locock procedure), and the red known, either. Pure grossular and pyrope have no
triangles plot the RI and susceptibility data. The var- transition elements in their composition and are thus
ious garnet compositions can be recognized by their diamagnetic. Their susceptibilities are very small and
position with respect to the end members. slightly negative; they were assigned by Hoover et al.
The process to determine the composition of the (2008) values of −0.2 × 10−4 SI, typical of diamagnetic
three garnet end members from any data point is materials. The other six end members are less easy to
simple. For example, for the green triangle represent- define. Frost (1960) measured the mass, or specific,
ing the Locock five end-member composition of susceptibilities of andradite, almandine, and spessar-
sample no. 234, which is plotted with an RI of 1.766 tine, which (when converted to volume susceptibili-
and a susceptibility of 23.7 × 10−4 SI: ty) are 23.8, 36.9, and 42.7 × 10−4 SI, respectively. But
the data are not robust. The four almandine-spessar-
• From the pyrope apex of the pyralspite ternary, tine garnets Frost measured, ranging from Alm65Sps35
draw a line through the center of the data point to Alm10Sps87, all had the same mass susceptibility.
to intercept the opposite base of the triangle, Nathan et al. (1965) measured spessartine’s volume
shown by the blue dashed line. susceptibility as 44.3 × 10−4 SI, but the author’s own
measurements on spessartine suggested that this was
• Next, measure the total length of the line, and slightly low. Hoover et al. (2008) were unable to find
then the length from the data point to the base measured susceptibilities for the other three garnet
of the triangle. end members. Approximate values, however, were
calculated based on the magneton numbers of the
• Divide the line length from the data point to
constituent transition element ions present (Kittel,
the base by the total length, which will give the
1956), using the Langevin equation. This is how
percentage of pyrope end member.
Hoover et al. (2008) obtained the values shown in the
For sample no. 234, the result is 47%. The pro- figures—30.8, 40.7, 47.5, 12.9, 6.9, and 13.7 × 10−4 SI,
cess can be repeated for the other two apices, but it respectively—for andradite, almandine, spessartine,
is simpler to measure the relative proportions of uvarovite, goldmanite, and knorringite.
almandine and spessartine on the almandine-spes-
sartine join where the blue line crosses it, and pro- Measured Properties. Except for the magnetic suscep-
portion them to the remaining percentage (53% for tibility measurements, all properties for the 28 study
this example). Here it is at 60% spessartine, which samples were supplied by GIA from the Manson-
yields 32% for spessartine and 21% for almandine, Stockton research. The volume susceptibility mea-
or Prp47Sps32Alm21. By comparison, the Locock tech- surements were taken by the author, using the appa-
nique characterizes this stone, rounding to the near- ratus described in box B, with cobalt chloride as a sus-
est 1%, as Prp43Sps34Alm18Grs4Adr1. ceptibility standard (described by Hoover et al., 2008).

End-Member Properties. The properties of each gar- RESULTS

net end member (table 1) are required to plot their RI vs. Magnetic Susceptibility Diagram. In the RIMS
positions in the various modified Winchell diagrams. diagram (figure 2), the gem ugrandites plot on or very
For every end member but knorringite, the RI, SG, near the line joining grossular and andradite. The
and cell constant used were reported by Meagher single mint green grossular (sample no. 198) is very
(1982). For knorringite, the calculated data from close to the grossular end member. The three hes-
McConnell (table 50 in Deer et al., 1982) were used sonites (2489, 27352, and 27423) are about 14%
by Hoover et al. (2008), but the data do not agree toward andradite. The two demantoids (2491 and
well with the RI and cell dimension data measured 13234) are close to the andradite end member. For


the pyralspite garnets, a mixed almandine-spessar- Comparing the measured and calculated end-
tine group (996, 5821, and 12588) plots along the member data for most samples shows fair agreement
almandine-spessartine join, quite distinct from the (figure 2 and table 2). The variations between the two
rest. Spessartine samples (12487, 13047, and 27257) techniques are about what one would expect due to
plot near their end-member composition. Three measurement error and some uncertainty in end-
stones, consisting of two malaya (234 and 491) and member properties. Not including sample 13113, for
one color-change garnet (79), are positioned within which the originally measured RI was incorrect, the
the pyralspite ternary. The rest of the garnets are average difference between measured and calculated
arrayed near the pyrope-almandine join, or within RI values is 0.003. For volume susceptibility, there is a
the grossular-pyrope-almandine ternary; three difference of 0.5 × 10−4 SI, if one disregards samples
chrome-pyropes (11089, 11090, and 13113) are 13047 and 26767, which are anomalous. For specific
included in the latter group. gravity (figure 3), the average difference is 0.01.




11588 5821

Figure 3. This plot shows

SG vs. magnetic suscep-
tibility (SGMS diagram)
4.05 for the 27 GIA garnets
for which the data were

available. However,
sample numbers are not
3.95 shown for those samples
that plot toward the
pyrope and grossular end
members, which overlap
Uv 2468 491
on this diagram.



3.65 End member

Grs Measured value
Locock-derived value

0 5 10 15 20 25 30 35 40 45 50 55



To give a sense of how these variations are (13047, 12487, and 27257) may also have been mea-
reflected in terms of garnet composition, let us sured incorrectly. Each had a reported RI of 1.800
assume a 3% difference in end-member composition from the Manson-Stockton data set, and each had a
of a Pyr50Alm50 garnet. This would produce a change very similar end-member composition. The calculat-
of 0.004 in RI, 1.22 × 10−4 SI in volume susceptibili- ed RIs are all 1.805. GIA, however, remeasured the
ty, and 0.02 in SG. Thus the average difference RI of sample 13047 as >1.810, or above the index of
between these three measured properties and those the refractometer liquid. The author checked these
calculated from the Locock procedure represents less values with an experimental deviation angle refrac-
than 3% compositional change in a mid-range tometer and found all three to be 1.809, but the error
pyrope-almandine. It is important to remember that range of the instrument is no better than 0.004.
such a derived composition will not be unique, and These stones remain problematic.
that any additional information, such as absorption
spectra, may further reduce uncertainties in deter- SG vs. Magnetic Susceptibility Diagram. Another
mining garnet composition. modified Winchell diagram, plotting specific gravity
The RIMS diagram shows the advantage of such vs. magnetic susceptibility (SGMS), holds some
plots in evaluating how accurate our property mea- promise. This diagram (figure 3) shows the 27 gar-
surements need to be. Consider the various joins in nets for which SG data were available. Stockton and
the illustration. Those between pyrope or grossular Manson (1985) did not consider specific gravity in
and the other end members are relatively long, and their characterization of garnet, though they mea-
therefore span a wide range in refractive index and sured it for many of the samples:
susceptibility. A difference of a few percentage
points in composition will have a measurable effect There is so much overlap in specific gravity ranges
on these properties. By comparison, the almandine- for the various types of garnets that the usefulness of
this property is questionable. Moreover the difficulty
spessartine join is quite short: a larger percentage
of accurately measuring density as well as the consid-
change in composition is needed to produce a mea- erable variability introduced by the presence of inclu-
surable change in properties. sions suggests that this is not a reliable characteristic
Regarding the two samples with anomalous sus- for the identification and classification of gems.
ceptibilities, no. 26767 had a small chip on the table (Stockton and Manson, 1985, p. 212)
below the magnet, which probably was responsible
for its lower susceptibility. Sample 13047, a spessar- However, figure 3 suggests otherwise. Note the
tine, had a high susceptibility. In the author’s collec- positions of samples 996, 11588, 26767, 26620, 3429,
tion, a Brazilian spessartine shows a similarly high 234, 491, 2211, and 2486 with respect to the pyral-
susceptibility. Neither shows evidence of ferromag- spite ternary in this figure, compared to their posi-
netic inclusions, such as magnetite, that could tions in figure 1. They indicate very similar chemical
explain their anomalous susceptibilities. The author composition, whether RI or specific gravity is used
suspects that these are yttrium-bearing spessartines, with magnetic susceptibility. Of particular impor-
with relatively enriched rare-earth contents that are tance for the practicing gemologist is that RI values
responsible for their high susceptibilities. Note that that cannot be obtained on a conventional refrac-
sample 13047, in this property space, falls well tometer can now be measured by substituting SG for
beyond the expected measurement error from any RI to infer composition in a manner similar to using
ternary triangle using the eight more-common gar- a RIMS plot. This possibility was suggested by
net end members. Hoover et al. (2008), assuming SG values were accu-
Chrome-pyrope 13113 originally had an anoma- rate enough. The Manson-Stockton data clearly
lously low measured RI, 1.732, compared to a calcu- demonstrate the usefulness of SG data for stones that
lated value of 1.744. Yet the other measured values are not too included. For example, metamorphic gar-
of SG, cell constant, and susceptibility agreed well nets of high almandine content can be easily distin-
with calculated values. The refractive index is clear- guished from typical pegmatitic almandine-spessar-
ly questionable, which illustrates the utility of tine compositions.
Winchell diagrams in checking the consistency Unfortunately, the SGMS plot also shows that
between measured properties and chemical data. the specific gravity and magnetic susceptibility of
The sample’s RI was rechecked by the author and pyrope and grossular are almost identical, and that
found to be 1.742. The RIs of the three spessartines the pyrope and grossular ends of the pyralspite and


ugrandite ternaries overlap. Clearly there are prob- an RISG diagram for 202 stones used in their studies,
lems in distinguishing a pyrope-almandine-spessar- but did not show the end-member positions. Notice
tine from a grossular-almandine-spessartine. It that chrome pyrope no. 13113 is again anomalous in
should be no surprise, then, that using SG to add a this diagram, but not in the SGMS plot of figure 3.
fourth end member to inferred compositions from RI There is a significant problem with using the
and susceptibility is questionable. RISG diagram: The pyralspite and ugrandite ternaries
are very narrow, indicating that within either, the
RI vs. SG Diagram. Figure 4 is the modified Win- relationship between the two properties is nearly lin-
chell plot of RI vs. SG (RISG), using the 27 stones ear. This is a consequence of the Gladstone-Dale rela-
shown previously. Jackson (2006) used such a dia- tionship between the two properties (Larsen and
gram for characterization, but noted the lack of Berman, 1934). Thus we are unable to distinguish—in
robustness for garnets of multiple end-member com- the case of a pyralspite, for example—the proportions
positions. Stockton and Manson (1985) also plotted of each of the three components based only on these





Figure 4. This plot shows
996 RI vs. SG (RISG diagram)

5821 12588 for the 27 GIA garnets

1.810 with data available.
11568 27257
1.800 Sps

13047 12487

1.760 2486
27352 2489 77
1.750 4967
1.740 25000
Grs End member
11090 Measured value
1.720 Locock-derived value

3.55 3.65 3.75 3.85 3.95 4.05 4.15 4.25 4.35



tions are 3.83 and 3.80, respectively, a difference of
less than 1%, which is very slight for such a large
difference in composition. The measured SG for this
1.88 ct stone is 3.85. Clearly, this value does not fit,
so one cannot calculate a mixed PrpSpsGrsAlm gar-
net from these measured data. For this stone, a
rather accurate SG (of 3.817) is needed to derive a
four end-member composition obtained with the
Locock method. This example demonstrates the
problem with adding SG to determine a fourth end
member for distinguishing between some pyralspite
garnets. A photo of sample no. 79 is shown in figure
5, along with no. 25000 and a hessonite that resem-
Figure 5. These garnets show similar color but have bles the latter in color.
different compositions. From left to right: GIA sam- Garnet no. 25000 is also an interesting stone.
ple nos. 79 (pyralspite), 37369 (hessonite), and 25000 From the Locock procedure, its composition is
(pyrope-almandine). Photo by Robert Weldon. Prp 68Alm 24Grs 3Adr 1. GIA had classified it as a
pyrope. Interestingly, the stone’s color matches that
two properties, unless we have extremely accurate of many hessonites. But its measured susceptibility
measurements of both properties. It is interesting to is close to its calculated properties, and the stone is
note that for these particular stones, the measured RI clearly not hessonite (figure 2). The measured prop-
and SG values agree reasonably well with the calcu- erties indicate Pyp75Alm25, if it is assumed to have
lated values. This is further evidence that Manson only pyrope and almandine components, and this
and Stockton were able to measure SG adequately to composition would have RI and SG values of 1.743
provide useful data for Winchell-type diagrams. and 3.77, respectively. However, this sample can
also be fit with a composition of Grs51Prp26Sps23 and
Multiple End Members and Compositional Possi- RI and SG values of 1.744 and 3.77, respectively.
bilities. A few garnet varieties, such as tsavorite and These properties fall within the measurement error
demantoid, are often characterized by a single end of Prp75Alm25. This example shows that very minor
member, but most other garnets require more. differences in measured properties may tip the com-
Figures 2 and 3 show that the malaya (nos. 234 and position to one side or the other of the pyrope-
491) and color-change (no. 79) garnets are the only almandine join. The author suspects there are other
pyralspites studied that plot well within the pyral- such pyropes that could masquerade as hessonites.
spite ternary. The others are either close to the
almandine-spessartine or pyrope-almandine joins, or RI vs. Cell Constant Diagram. While Manson and
within the grossular-pyrope-almandine ternary. The Stockton did not use cell constant in their articles,
malaya and color-change garnets are best described they did measure it for a number of their samples.
by a combination of the pyralspite end members, This makes it possible to compare the modified
plus significant grossular. Thus they are very useful Winchell RIMS plot against a standard Winchell
for comparing the inferred three or four end mem- plot of RI vs. cell constant (RICC), as well as four
bers by applying Winchell diagrams to the micro- other variations, when four quantitatively measur-
probe composition. able properties are available.
Sample no. 79 is particularly useful because of Figure 6 shows a conventional Winchell plot for
its relatively high grossular component. The Locock 16 of the 28 GIA samples for which cell constant
procedure yields a composition of Prp48Sps23Grs15 was measured. Note that the differences between
Alm12Uv1, with less than 1% consisting of other the calculated and measured values are about the
components. From figure 2 the composition can be same as those shown for the RIMS plots in figure
derived as (1) Prp63Sps20Alm17 or (2) Sps35Grs34Prp31. 2. Comparing these two plots shows distinct dif-
Either of these compositions gives an RI of 1.751 ferences in the positioning of some stones within
and a susceptibility of 16.4 × 10 −4 SI. Note that nei- the pyralspite ternaries.
ther PrpGrsAlm nor GrsAlmSps fits the two proper- Consider the stones arrayed near the pyrope-
ties. The calculated SGs from these two composi- almandine join. In figure 2 they are positioned, on








Figure 6. This con-
1.820 ventional Winchell
plot shows the RI vs.

996 cell constant (RICC

diagram) for 16 GIA
Sps garnet samples.


1.770 2211
77 79

End member
1.720 Measured value

Locock-derived value

11.4 11.5 11.6 11.7 11.8 11.9 12.0 12.1


average, slightly to the left of the join (nos. 26767, Winchell diagram used. In these cases, the differ-
3429, 2211, 2486, 77, and 25000). In figure 6 they are ences are due to grossular and/or uvarovite compo-
within the pyralspite ternary. But the malaya (no. nents. Note that in the RICC plot the ugrandite
491) and color-change (79) garnets are outside the ternary is very narrow, indicating that a ugrandite
pyralspite ternary. These differences reflect the with all three end members as major components
effects of end members other than pyralspite’s on the cannot be well characterized by this type of diagram.
measured properties. If the composition of a garnet is
purely pyralspite, its position within the ternary Usefulness of Modified Winchell Diagrams. These
would not change according to the particular diagrams provide a simple but powerful demonstra-


tion of the relationship between garnet properties types. The RIs and susceptibilities were calculated
and composition. Any garnet with properties that from Wright’s average compositions. The plot shows
plot near an end member will have a preponderance that garnets associated with felsic igneous rocks
of that end member in its composition. A sample (nos. 1, 2, and 3) such as the granitic pegmatites are
positioned toward the center of a given ternary will essentially almandine and spessartine. Metamorphic
be composed of similar quantities of each end mem- garnets (nos. 4, 5, and 6) show higher pyrope/grossu-
ber. One can also estimate the effect that adding or lar as the metamorphic grade increases (blue arrows),
subtracting any end member will have on the various with the highest grade (eclogitic; no. 6) approaching
properties. The effect on properties is directly related the composition of peridotitic garnet (no. 7). The red
to the garnet’s distance from the end member. line shows the change in properties of an average
In addition to offering compositional informa- peridotitic pyrope toward the knorringite compo-
tion, the modified Winchell diagrams can also pro- nent as chromium is added. Comparing figure 7
vide insights into garnet paragenesis. Figure 7 is a with figure 2, correlations are evident between many
RIMS plot of data from Wright (1938) showing the of the garnet samples and their probable genetic ori-
average compositions of garnets from various rock gins. Malaya and color-change garnets are not repre-
sented in the Wright data, but the present author
suspects they represent metamorphosed, subducted,
high-Mn oceanic sediments.
Figure 7. This RIMS diagram shows the average gar-
net compositions from various rock associations, as
given by Wright (1938). The RI and susceptibility CONCLUSION
data are plotted from compositions that include the
Gem garnets (e.g., figure 8) encompass a broad range
average proportions of five major garnet end mem-
bers. The blue arrows show the compositional
of compositions and properties. Because most
change for metamorphic garnets as the metamor- gemologists lack the capability to obtain quantita-
phic grade increases. tive chemical data, garnet composition must be
inferred from measured/observed properties. In the
past, RI and possibly SG were the only quantitative
RIMS DIAGRAM measures for deriving chemical composition. Yet
these properties are not sufficiently independent of
1.890 Adr each other for such determinations. By measuring
1.880 Kn magnetic susceptibilities on selected garnets with
1.870 Uv well characterized compositions, the author has
1.860 demonstrated a technique for inferring garnet com-
1.850 position from measurements of two or more quanti-
1.840 Go
tative properties.
Alm Inferring chemical composition in this way
should not be considered equivalent to the results

1.820 2
1 obtained from a microprobe or other chemical anal-
ysis. While chemical data typically give the percent-
Sps ages of oxides in a sample, these data are not of
1.790 8
prime interest to the gemologist, who seeks the pro-
1.780 5
portion of ideal end members. This proportion may
be obtained from either oxide chemistry or garnet
1. Pegmatites
properties, but neither method gives unique results.
1.750 2. Granites
3. Contact-altered Si-rich Rocks
When oxide percentages are used, the number of
1.740 6 4. Biotite Schists
5. Amphibole Schists
ideal end members will vary according to the num-
Grs 7
1.730 6. Eclogites ber of elements analyzed. Using garnet properties to
7. Kimberlite & Peridotites
1.720 8. Various Basic Rocks infer end-member composition limits the number
9. Calcareous Contact Rocks
of end members to one more than the number of
0 5 10 15 20 25 30 35 40 45 50
properties measured. Using RI and magnetic suscep-
VOLUME SUSCEPTIBILITY (x10-4 SI) tibility, then, we can infer a three end-member


The various modified Winchell diagrams give
the gemologist new insight into garnet chemical
composition and its relation to measured properties.
They can be a useful educational tool—showing the
range of RIs possible where pyrope is the principal
component, for example. In addition, Winchell dia-
grams can yield information on a garnet’s probable
geologic environment.


Dr. Hoover ( is a retired research
geophysicist with the U.S. Geological Survey who now
pursues independent gemological research in Spring-
field, Missouri.

The author thanks GIA for providing access to the
enormous amount of data collected in the 1980s by D.
Vincent Manson and Carol M. Stockton for their classic
series of Gems & Gemology articles on the garnet
group. Particular thanks go to Terri Ottaway, museum
Figure 8. Gem garnets, such as these examples from the curator, for supplying samples from GIA’s Museum
Dr. Edward J. Gübelin Collection, show a variety of prop- Collection in Carlsbad, and to Ms. Stockton for provid-
erties and compositions. Magnetic susceptibility is one ing her raw data. Without their cooperation, this work
more measureable property the gemologist can use to help would not have been possible. The author is also
estimate garnet composition. Shown clockwise from the grateful to Bear and Cara Williams (Bear Essentials,
top: 44.28 ct pyrope-spessartine from Madagascar (GIA Jefferson City, Missouri) and Claire Mitchell (Gem-A,
Collection no. 34387a), 19.12 ct pyrope-almandine from London) for innumerable discussions on the nature and
Sri Lanka (34769), 4.24 ct demantoid (33303), 3.65 ct tsa- identification of the garnet group.
vorite from Tanzania (35569), and 19.90 ct spessartine
from Brazil (33238); photo by Robert Weldon.

REFERENCES j.cageo.2007.12.013.
Deer W.A., Howie R.A., Zussman J. (1982) Rock Forming Minerals, Manson D.V., Stockton C.M. (1981) Gem garnets in the red-to-
Vol. 1A—Orthosilicates. Longman, London, pp. 469–697. violet color range. G&G, Vol. 17, No. 4, pp. 191– 204,
Frost M.J. (1960) Magnetic susceptibility of garnets. Mineralogical
Magazine, Vol. 32, pp. 573–576, Meagher E.P. (1982) Silicate garnets. In Reviews in Mineralogy,
minmag.1960.032.250.07. Vol. 5, 2nd ed., Orthosilicates, Mineralogical Society of
Hoover D.B., Williams B. (2007) Magnetic susceptibility for gem- America, Washington, DC, pp. 25–66.
stone discrimination. Australian Gemmologist, Vol. 23, No. 4, Nathan Y., Katz A., Eyal M. (1965) Garnets from the Eilat area,
pp. 146–159. southern Israel. Mineralogical Magazine, Vol. 35, pp. 386–392,
Hoover D.B., Williams C., Williams B., Mitchell C. (2008)
Magnetic susceptibility, a better approach to defining garnets. Nixon P.H., Hornung G. (1968) A new chromium garnet end
Journal of Gemmology, Vol. 31, No. 3/4, pp. 91–103. member, knorringite, from kimberlite. American Miner-
Hutchison C.S. (1974) Laboratory Handbook of Petrographic alogist, Vol. 53, pp. 1833–1840.
Techniques. John Wiley & Sons, New York, 527 pp. Rickwood P.C. (1968) On recasting analyses of garnet into end-
Jackson B. (2006) The garnets. In M. O’Donoghue, Ed., Gems, 6th member molecules. Contributions to Mineralogy and Petrology,
ed., Elsevier, Oxford, UK, pp. 195–237. Vol. 18, pp. 175–198,
Kittel C. (1956) Introduction to Solid State Physics, 2nd ed. John Ringwood A.E. (1977) Synthesis of pyrope-knorringite solid solu-
Wiley & Sons, New York, 617 pp. tion series. Earth and Planetary Science Letters, Vol. 36, pp.
Larsen E.S., Berman H. (1934) The microscopic determination of 443–448,
the nonopaque minerals. U.S. Geological Survey Bulletin 848, Stockton C.M., Manson D.V. (1985) A proposed new classification
2nd ed., U.S. Government Printing Office, Washington, DC, for gem-quality garnets. G&G, Vol. 21, No. 4, pp. 205–218,
266 pp.
Locock A.J. (2008) An Excel spreadsheet to recast analyses of gar- Winchell H. (1958) The composition and physical properties of
net into end-member components, and a synopsis of the crys- garnet. American Mineralogist, Vol. 43, pp. 595–600.
tal chemistry of natural silicate garnets. Computers & Geo- Wright W.I. (1938) The composition and occurrence of garnets.
sciences, Vol. 34, pp. 1769–1780, American Mineralogist, Vol. 23, pp. 436–449.


Ron H. Geurts, Ilene M. Reinitz, Troy Blodgett, and Al M. Gilbertson

ations during planning and cutting. Likewise, mak-

ers of non-contact optical scanners have been inter-
Grade boundary limits are presented for 10
ested in guidelines for how measurable symmetry
symmetry parameters of the round brilliant cut
parameters affect the GIA symmetry grade. The
diamond. Starting in early 2012, these values
grade boundaries presented here offer a substantive
will be used to support and constrain visual
estimate of the symmetry grade for any round bril-
symmetry grading on GIA diamond reports.
liant cut diamond.
For manufacturers, the boundaries provide
In GIA’s laboratory, polished diamonds are mea-
useful predictions of symmetry grades. Other
sured with a non-contact optical scanner early in the
symmetry features of faceted diamonds will
grading process. Later, polish and symmetry are eval-
continue to be evaluated visually.
uated visually at 10× magnification, using a standard
procedure. As described in Gillen et al. (2005), specif-
ic parameter- and facet-related features are consid-
ered in grading symmetry. This article presents
numerical grade limits for 10 important symmetry

ince 2006, GIA has used certain proportion mea- parameters that can be measured accurately enough
surements obtained with non-contact optical to support visual symmetry grading. Although mea-
scanners to grade the cut of round brilliant dia- sured values have been available to graders as a guide
monds. Improvements in the operation and accuracy for several years, beginning in 2012 GIA will use
of these instruments now enable us to also measure measured values and apply these boundary limits
some symmetry parameters during the grading pro- strictly when grading symmetry for round brilliant
cess. Although both Excellent and Very Good sym- cut diamonds. Facet-related symmetry features, and
metry grades meet GIA’s criteria for an Excellent cut the manner in which multiple features combine,
grade (Moses et al., 2004), there is a premium for may also affect the symmetry grade, and these
what the trade calls a “triple Excellent”: an aspects will continue to be evaluated visually, as
Excellent grade for cut, polish, and symmetry. they are presently beyond reproducible instrument
Therefore, many diamond manufacturers would like measurement.
to be able to predict GIA symmetry grades from Compared to visual assessment, instrumental
measurement data, so they can apply these consider- measurements provide a more consistent way of
establishing a symmetry grade, especially when a dia-
mond has very subtle symmetry deviations. Figure 1
shows a diamond with several symmetry flaws: a
See end of article for About the Authors.
GEMS & GEMOLOGY, Vol. 47, No. 4, pp. 286–295,
wavy and uneven girdle (resulting in an uneven crown height), a table not parallel to the girdle, and
© 2011 Gemological Institute of America uneven bezel facets. In the past, the only means of

Figure 1. This 0.69 ct
standard round bril-
liant cut diamond
displays several
obvious symmetry
features that can be
quantified. Photo by
Robert Weldon.

determining this diamond’s symmetry grade would measure crown angle to three decimal places, but if
have been the judgment and experience of the grader. repeated measurements demonstrate an uncertainty
Quantifying these features by instrumental measure- in the first decimal place, the two additional digits
ment provides a more consistent basis for symmetry offer no meaningful precision (Reinitz et al., 2005).
grading, and gives cutters the details needed to Even detailed knowledge of the uncertainty does
improve their diamonds’ symmetry. not tell us whether measurements are accurate.
Accuracy can only be determined relative to the mea-
BACKGROUND surement of a known standard, such as an object
The repeated measurement of any attribute, such as with NIST-traceable values and reported uncertain-
weight or size, is accompanied by a certain degree of ties. Box A describes some basic metrology concepts,
uncertainty. For example, the repeated measurement including measurement uncertainty.
of a diamond’s weight, or its total depth, yields The proportion values used to determine a dia-
results that vary slightly within a certain range. For mond’s cut grade are normally the average of eight
the most accurate results, the measured value itself measurements; these are not greatly affected by a sin-
and the variation in repeated measurements—the gle outlying value. In contrast, symmetry parameters
uncertainty of that value—are both important. examine the range (the largest and smallest) of those
The U.S. National Institute of Standards and values, and they are much more affected by a single
Technology (NIST) notes that “a measurement result poor measurement. This means a higher level of con-
is complete only when accompanied by a quantita- fidence in the reproducibility and accuracy of each
tive statement of its uncertainty” (“Uncertainty of measured value is needed to predict a symmetry
measurement results,” 1998). Whatever the tool or grade, or to reinforce visual grading. GIA has
method, measurement results fall within a certain achieved this confidence through advances in the
allowable range of values—the tolerance. For our pur- devices used to measure polished diamonds, coupled
poses, the tolerance of a measuring device describes with efforts to ensure the diamonds are thoroughly
its contribution to the overall uncertainty of the mea- cleaned. For example, suppose eight crown angles are
sured values (GIA Research, 2005). individually measured at 34.1°, 34.5°, 34.9°, 35.3°,
Statistical examination of repeated independent 34.2°, 34.3°, 34.0°, and 34.1°. The average is
measurements provides one way to estimate their 34.43°, and the difference in values (maximum
uncertainty. The distribution of these measurements minus minimum) is 1.3°. A second set of measure-
also reveals information about reproducibility and ments yields values of 34.1°, 34.5°, 34.5°, 34.8°, 34.2°,
defensible precision. For example, a device might 34.3°, 34.0°, and 34.1°. The second average is 34.31°,

of the eight individual values. In a round brilliant of
T aking several independent measurements of the
same characteristic illustrates the difference
between precision and accuracy, as shown in figure
lower symmetry, the eight crown angle values may vary
by several degrees. The uncertainty of a symmetry
A-1. Accuracy refers to how close the measured val- assessment for such variation among the crown angles
ues are to the reference value, shown here as the
center of the target. Precision refers to how close the
values are to each other, and in practice this affects
Figure A-1. A measurement is accurate when it
how many significant figures should be used when agrees with an independently obtained reference
reporting the measurement. value (here, the center of the bull’s-eye). Measure-
When the difference between two measured val- ments are precise when they can be reproduced with
ues is less than or equal to the measurement uncer- small uncertainties. The ideal situation is to have
tainty, the values are within tolerance of each other, measurements that are both accurate and precise.
and by definition not readily distinguishable from
one another. Figure A-2 shows six measurements of
the total depth of one round brilliant, each with an
uncertainty of ±0.015 mm. The average value of
those measurements is 5.015 mm. Trial 4, with a
value of 5.00 mm, is just within tolerance of that
average. Trial 6, with a value of 5.04 mm, is not
within tolerance of the average. This is described as
an outlying value. Not Accurate
Not Precise
Not Precise
Most gemologically important parameters for the
round brilliant cut diamond, such as the crown or
pavilion angle, represent averages rather than single
measurements. In metrology, averages of multiple
measurements are used to reduce measurement
uncertainty. But a quantity such as average crown
angle is calculated from eight values obtained from
different facets, rather than eight measurements of
Not Accurate Accurate
the same facet. As a result, this average has its own Precise Precise

uncertainty that is no smaller than the uncertainties

only 0.12° below the previous average. But the differ- grade boundary. Multiple measurements, on one
ence in values is now 0.8°, considerably smaller than device or on different devices, can yield slightly dif-
the 1.3° from the first set of measurements. In other ferent results. All devices have a margin of error
words, the average changes less than the difference in (within the tolerance of the device) that could yield
values when one or two of the eight values is marred two different grades when one or more parameters
by dirt or some other measuring problem not specifi- are near a border. Since no measurement is exact,
cally related to that particular diamond. prudent cut planning acknowledges measuring toler-
Higher-quality measurements have a smaller ances and avoids placing parameters too close to the
uncertainty, but even the best measuring systems borders.
have some tolerance for each parameter. Box A
shows an example of measurement uncertainty at
the border between Very Good and Excellent. Even MEASURABLE SYMMETRY PARAMETERS AND
measurements of clean diamonds made on devices of ADDITIONAL FACTORS
proven accuracy and precision can produce one or Ten symmetry parameters are illustrated in figure 2.
more values that fall within tolerance of a symmetry GIA has developed procedures to measure these




5.03 1.3

5.02 1.2

5.01 1.1
5.00 1.0
4.99 0.9
4.98 0.8
1 2 3 4 5 6 1 2 3 4 5
Figure A-2. These total-depth measurements are shown
with error bars that represent measurement uncertainty.
Figure A-3. A round brilliant measured five times
These bars overlap the average value of 5.015 mm for the
yields crown angle–variation values with uncertain-
first five trials, but not the sixth. It is important to recog-
ties that cross the symmetry grade limit for this
nize the distinction between (1) measurements within
parameter (1.2°). Although the third measurement of
tolerance of each other, and (2) measurements that clear-
1.3° would indicate Very Good symmetry, the most
ly differ from each other beyond the tolerance. If the error
reproducible value—the one most often obtained—is
bars overlap each other, the measurements can be consid-
within the limits for Excellent.
ered the same; if they do not overlap, the measurements
are different.

is also no smaller than the individual uncertainties. into two differing results. Figure A-3 shows such an
The uncertainty associated with a measured example, where all five measurements are within tol-
value can be thought of as a “bubble” around it. erance of each other, but one generates a symmetry
Overlap among these bubbles in a group of measure- grade of Very Good, based on this one parameter,
ments indicates agreement with each other. A fixed while the other four would score in the Excellent
boundary, such as a limit for symmetry grading, can range. From basic metrological principles, if the mea-
cut through such uncertainty bubbles, separating a suring device is sound, the more reproducible value
group of measurements that agree with each other is the correct one.

parameters with sufficient accuracy to determine

the symmetry grade of round brilliant cut diamonds. NEED TO KNOW
Additional parameters have been identified, such as
• Starting in early 2012, GIA will apply boundary
the symmetry of the star facets and the upper and
limits for 10 symmetry parameters measured by
lower girdle facets, but numerical boundaries for non-contact optical scanners when grading the
these are still under review. symmetry of round brilliant cut diamonds.
The 10 symmetry parameters are calculated as
• Additional measurable parameters, aspects arising
follows: from combinations of these parameters, and
1. Out-of-round: the difference between the maxi- facet-related symmetry variations will continue to
mum and minimum diameter, as a percentage be assessed visually.
of the average diameter • Manufacturers should strive to attain values that are
2. Table off-center: the direct distance between 20% lower than the symmetry boundary limits, to
the table center and the outline center projected account for measurement uncertainty and features
into the table plane, as a percentage of the aver- that may combine to lower the symmetry grade.
age diameter


Table off-center:
Out-of-round: deviation
deviation of the table
from the circular shape
from the central
of a round diamond
position on the crown

Culet off-center:
deviation of the culet from Table/culet alignment:
the central position on the displacement of the
pavilion table facet and culet in
opposite directions

Crown height variation: differing crown Crown angle variation: crown angles are
height measurements indicating a wavy unequal; typically related to table off-center
girdle or table/girdle not parallel

Pavilion depth variation: differing pavilion depth Pavilion angle variation: pavilion angles are
measurements indicating a wavy girdle unequal; typically related to culet off-center

Girdle thickness variation: variation of the

girdle thickness at bezel positions
Table size variation:
differing table size
measurments indicating
non-octagonal table

Figure 2. These 10 symmetry features can be measured reliably enough by non-contact optical scanners to
determine the symmetry grade of round brilliant cut diamonds.

Figure 3. Three vertical lengths (A1–A3)
in this close-up of the 0.69 ct diamond in
figure 1 illustrate girdle thickness differ-
ences. Region B (green circle) shows
where the facet edges of the upper and
lower girdle do not meet (crown and
pavilion misalignment). Region C (yel-
B low circle) shows the junction where
three facets fail to meet (pointing
fault). Photo by Robert Weldon.

3. Culet off-center: the direct distance between interaction plays a large role in determining the
the culet center and the outline center project- overall symmetry grade for round brilliants with
ed into any horizontal plane such as the table lower symmetry. But for those with high symmetry,
plane, as a percentage of the average diameter the magnitude of a single feature may dominate the
4. Table/culet alignment: the direct distance evaluation.
between the table center and the culet center Facet-related symmetry features also play a role in
projected into the table plane, as a percentage of determining the symmetry grade (e.g., figure 3, fea-
the average diameter tures B and C), but they are not part of the grading pro-
5. Crown height variation: the difference between cedure described here. A full description of facet-relat-
the maximum and minimum crown height val- ed symmetry features can be found in Blodgett et al.
ues, as a percentage of the average diameter (2009). Open or short facets (non-pointing), misalign-
ment between the bezels and pavilion mains, and
6. Crown angle variation: the difference between
prominent naturals or extra facets are readily
the maximum and minimum crown angle val-
observed, but they may occur independently of the 10
ues, in degrees
measurable symmetry parameters listed above.
7. Pavilion depth variation: the difference Misshapen or uneven facets usually relate to a combi-
between the maximum and minimum pavilion nation of the 10 parameters, but the relationships can
depth values, as a percentage of the average be complex.
8. Pavilion angle variation: the difference between RECOMMENDED SYMMETRY BOUNDARIES
the maximum and minimum pavilion angle The limits given below were derived from a statisti-
values, in degrees cal comparison of measured values for the 10 param-
9. Girdle thickness variation: the difference eters and the final symmetry grades assigned to the
between the maximum and minimum girdle diamonds. This comparison was repeated four times
thickness values, as a percentage of the average over a period of 10 years, each time on newly
diameter, measured at the bezel-main junctions acquired data sets from several thousand diamonds.
(see also features A1–A3 in figure 3) Each analysis examined several sets of limits for the
10. Table size variation: the difference between the 10 parameters to identify robust matches with visual
maximum and minimum table size values, as a symmetry grading.
percentage of the average diameter Table 1 presents the ranges of allowed values for
individual symmetry features, measured in percent-
Because the facets of a round brilliant are con- age or degrees, that GIA uses to support and con-
nected to each other, these symmetry features fre- strain visual symmetry grading. The limits dividing
quently occur in combination. All of the symmetry Fair from Poor symmetry are not presented here
features combine to produce a general face-up visual because of the small number of round brilliants with
impression, so the symmetry grade is established by such low symmetry. Measured values should be
looking at the face-up diamond. Depending on rounded to the indicated precision, if necessary,
where they occur, and how they combine, different before calculating the differences. If the value for any
symmetry features can visually amplify or compen- one parameter falls into a range associated with a
sate for one another, as discussed in box B. This lower grade, the overall symmetry grade will be low-


T he red dashed lines in the drawings in figure B-1

show the position of the table center, and the
blue dot shows the center of the stone outline.
ness are added to the off-center table and culet, the
visual difference between various combinations of
these features becomes even more pronounced. In
Consider two round brilliants, each with a 4% off- figure B-2 (top), the table and culet are off center in
center table and culet (cases B and C). The measured the same direction, and the girdle and crown height
values for table and culet being off center are equal, are uneven along this same A-B axis. Arranged in
and each feature would be easily noticed individual- this way, these features tend to compensate each
ly, in profile as well as face-up. When the culet and other visually, particularly in the face-up view. In
table are off center in the same direction (case B), the contrast, figure B-2 (bottom) shows a table and
two symmetry features compensate for each other culet that are off center in opposite directions, and
visually. But when the table and culet are shifted off the girdle thickness and crown height are uneven
center in different directions (case C), the negative in a different direction (along the G-H axis). This
visual impression is amplified considerably. combination amplifies the visual impression of
When uneven crown height and girdle thick- asymmetry.

Figure B-1. A culet and A B C

table that are off center in
different directions produce
a more asymmetrical
appearance (case C) than
when they are off center in
the same direction (case B).
Note that the degree of
asymmetry is extreme,
down to the Fair range. Culet and table centered Culet and table off center Culet and table off-center
4% in same direction 4% in opposite direction

ered accordingly. Combinations of symmetry fea- TABLE 1. Limits used by GIA to grade the symmetry of
tures, as well as facet-related features that are not round brilliant cut diamonds.
measured, will still be evaluated visually, which Parameter Excellent Very Good Good
may also contribute to a lower symmetry grade.
Out-of-round (%) 0–0.9 1.0–1.8 1.9–3.6
For example, if nine of the parameters are within Table off-center (%) 0–0.6 0.7–1.2 1.3–2.4
the Excellent range but the table is off-center by Culet off-center (%) 0–0.6 0.7–1.2 1.3–2.4
0.7%, the best possible symmetry grade is Very Table/culet alignment (%) 0–0.9 1.0–1.8 1.9–3.6
Good. If all 10 parameters are within the Excellent Crown height variation (%) 0–1.2 1.3–2.4 2.5–4.8
range, the expected symmetry grade would be Crown angle variation (°) 0–1.2 1.3–2.4 2.5–4.8
Excellent. But consider a round brilliant that is out Pavilion depth variation (%) 0–1.2 1.3–2.4 2.5–4.8
of round by 0.7%, with crown angle variation of 1.1° Pavilion angle variation (°) 0–0.9 1.0–1.8 1.9–3.6
and girdle thickness variation of 1.1%. Even though Girdle thickness variation (%) 0–1.2 1.3–2.4 2.5–4.8
all three parameters are within the limits for Table size variation (%) 0–1.2 1.3–2.4 2.5–4.8

Figure B-2. These two round brilliants have multiple measurable symmetry faults that limit them to
no better than a Good symmetry grade. Although both stones have equal culet off-center values, the
appearance of overall symmetry is different because of the relative placement of the various symmetry
faults. The green crosshair indicates the center of the outline, blue is the center of the table, and red
denotes the center of the culet. When the faults are aligned, the asymmetry appears less pronounced
(top). By comparison, when symmetry faults occur in different directions, the visual impression of
asymmetry is amplified (bottom). In either combination, these displacements are considerably more
subtle than those shown in figure B-1.

TABLE 2. Recommended limits for estimating the Excellent, the combination of these three symmetry
symmetry grade of round brilliant cut diamonds. features (and any others found on the diamond) may
Parameter Excellent Very Good Good result in either an Excellent or a Very Good symme-
Out-of-round (%) 0 – 0.7 0.8–1.4 1.5–2.8
try grade, depending on the visual assessment.
Table off-center (%) 0 – 0.5 0.6–1.0 1.1–1.9 Because every measurement contains uncertain-
Culet off-center (%) 0–0.5 0.6–1.0 1.1–1.9 ty, and symmetry features may combine to lower
Table/culet alignment (%) 0–0.7 0.8–1.4 1.5–2.8 the symmetry grade, we recommend a “safety mar-
Crown height variation (%) 0–1.0 1.1–2.0 2.1–3.9 gin” for the trade to use in estimating the symmetry
Crown angle variation (°) 0–1.0 1.1–2.0 2.1–3.9 grade. Accordingly, the values shown in table 2 are
Pavilion depth variation (%) 0–1.0 1.1–2.0 2.1–3.9 20% lower than those in table 1. When the values
Pavilion angle variation (°) 0–0.7 0.8–1.4 1.5–2.8 for all 10 parameters fall within these narrower rec-
Girdle thickness variation (%) 0–1.0 1.1–2.0 2.1–3.9 ommended borders, there is a strong likelihood that
Table size variation (%) 0–1.0 1.1–2.0 2.1–3.9
the visual symmetry assessment will agree with the

Figure 4. These three
round brilliants each dis-
play a combination of
symmetry faults. (A) The
table of this 1.00 ct dia-
mond (Fair symmetry) is
not an octagon (6.1%
table size variation, as
shown by the blue and
yellow lines) and the table
is off-center by 2.5%. The
asymmetry of the table is
associated with crown
angle variations and
uneven bezels (marked
red). (B) The culet of this
0.83 ct diamond (Fair
symmetry) is off-center by
2.9% (red dot). The table
is also off-center in an
opposing direction (green
dot), yielding a value for
table/culet alignment of
3.4%. These symmetry
faults are associated with
uneven bezels (marked
red) and pavilion mains.
Unlike the diamond in A,
the nearly equal quad-
rants defined by the yel-
low lines show that the
table is octagonal. (C) In
this 0.69 ct diamond (also
shown in figures 1 and 3;
Good symmetry), the gir-
dle is wavy and not paral-
lel to the table. Photos by
GIA (A and B) and Robert
Weldon (C).

measurement. Within these recommended limits, it though, it is more likely that multiple symmetry
is unlikely that a combination of measurable sym- features will limit the grade, because the interac-
metry features would lead to a lower symmetry tions among symmetry factors become more pro-
grade. Note that the second example in the previous nounced (again, see box B). Because combinations of
paragraph exceeds two of these recommended limits. minor symmetry features can create a significant
The boundary values presented for these 10 sym- visual impact, the limits in the tables must be
metry features are most useful along the viewed only as a guide.
Excellent–Very Good symmetry border, where a sin-
gle feature often dominates the final grade determi- DISCUSSION
nation. These individual parameter limits are also During the analysis of laboratory grading results, we
relevant for the border between Very Good and observed some variation in how strictly symmetry
Good. When symmetry problems become severe, was evaluated by our graders, particularly for mea-

sured features near the border between Excellent and emphasizes the visual impact of the off-center
Very Good. A common set of fixed numerical limits culet (again, see box B), but the measured values—
for these parameters can only improve the consisten- that is, Good for table-culet alignment, but Fair for
cy of symmetry grading for such stones. Diamonds table off-center—provide a context for evaluating
with at least one parameter beyond the limits shown the severity of the combination. In figure 4C, the
in table 1 will receive the lower symmetry grade. most prominent symmetry fault is displayed for
Symmetry features not captured by these 10 param- the diamond shown in figure 3. The table and gir-
eters will continue to be evaluated visually. If these dle are not parallel, a fault that is more severe than
additional facet-related features are sufficiently the uneven girdle thickness or the facet-related
prominent—an extra facet polished at the corner of symmetry features.
the table, for instance—they will reduce the sym-
metry grade even if all measured parameters fall
within the narrower limits in table 2. Visual sym- CONCLUSION
metry observations cannot raise a symmetry grade, Measurement is a process full of inherent uncertain-
but they can reveal instances when a cleaner, more ties, but GIA’s efforts to achieve smaller uncertain-
correct measurement of the diamond is needed. ties have been successful. Starting in early 2012, the
Measured values can be of great help for dia- measurable values presented in table 1 will be used
monds with multiple symmetry faults, such as the to attain greater consistency than is possible
three shown in figure 4. In such cases, some of the through visual assessment alone. Additional mea-
symmetry features are more easily noticed visual- surable parameters, aspects arising from combina-
ly, while others are captured more accurately by tions of these parameters, and facet-related symme-
measurement. In figure 4A, the asymmetry of the try variations will continue to be assessed visually.
table leads to variation in crown angles and A more restrictive set of limits is recommended for
uneven bezel facets. In other cases, similar faults manufacturers, to help ensure that the final symme-
with the table might be associated with a wavy try grade will not be undermined by combination
girdle that takes up the uneven aspects of the crown effects or measuring tolerances.
and allows little variation in the crown angles. Under
both sets of circumstances, the uneven bezels are a
prominent feature that does not describe the underly-
ing symmetry faults as clearly as the measured values Mr. Geurts is a manager of research and development at GIA
for crown angle variation, crown height variation, and in Antwerp. Dr. Reinitz is a project manager at GIA in New
girdle thickness variation. York. Dr. Blodgett is a research scientist, and Mr. Gilbertson a
In figure 4B, the off-center culet and table lead research associate, at GIA in Carlsbad.
to uneven bezels and pavilion mains. The displace-
ment between the table center and the culet

REFERENCES Moses T.M., Johnson M.L., Green B., Blodgett T., Cino K., Geurts
Blodgett T., Geurts R., Gilbertson A., Lucas A., Pay D., Reinitz I., R.H., Gilbertson A.M., Hemphill T.S., King J.M., Kornylak L.,
Shigley J., Yantzer K., Zink C. (2009) Finish, culet size and gir- Reinitz I.M., Shigley J.E. (2004) A foundation for grading the over-
dle thickness; Categories of the GIA Diamond Cut Grading all cut quality of round brilliant cut diamonds. G&G, Vol. 40,
System. No. 3, pp. 202–228,
girdle_highres.pdf [date accessed: June 14, 2011]. Reinitz I., Yantzer K., Johnson M., Blodgett T., Geurts R.,
GIA Research (2005) Measurement tolerances: Accuracy and pre- Gilbertson A. (2005) Proportion measurement: Tolerances for
cision in the gem industry. Rapaport Diamond Report, Vol. the GIA Diamond Cut Grading System. Rapaport Diamond
28, No. 13, pp. 183–185, Report, Vol. 28, No. 30, pp. 34–39,
RDR_pg183_185.pdf. pdf/0805_pg34_39.pdf.
Gillen D.B., Lanzl B.F., Yantzer P.M. (2005) Polish and symmetry. Uncertainty of measurement results (1998) The NIST Reference
Rapaport Diamond Report, Vol. 28, No. 39, pp. 80–87, on Constants and Uncertainty. Uncertainty/international1.html [date accessed: June 14, 2011].



Michael S. Krzemnicki, Franz Herzog, and Wei Zhou

Cu(Al,Fe 3+)6(PO4)4(OH)8• 4H2O, has been known

since prehistoric times. It has been widely used in
A set of six antique brooches, set with diamonds jewelry in the Middle East (Egypt and Persia), the Far
and light blue cabochons, was investigated with East (Tibet, Mongolia, and China), and by native
North Americans (Ahmed, 1999; Chalker et al.,
microscopy, Raman analysis, and EDXRF spec-
2004). Yet turquoise was once very fashionable in
troscopy. Most of the cabochons proved to be Europe, especially during the 18th and 19th cen-
fossilized dentine, also known as odontolite turies (Bennett and Mascetti, 2003), so it is not sur-
(mineralogically, fluorapatite). The brooches prising that imitations were used when genuine
also contained turquoise and artificial glass. turquoise was not available. The wide range of
turquoise imitations includes secondary minerals
from copper deposits such as chrysocolla, dyed min-
erals such as magnesite or howlite, and artificial
materials such as glass or sintered products (Arnould
and Poirot, 1975; Lind et al., 1983; Fryer, 1983; Kane,

he Swiss Gemmological Institute SSEF recent- 1985; Hurwit, 1988; Salanne, 2009).
ly received a set of six antique brooches for In this study, we report on a historic turquoise
identification (figure 1). These same pieces had substitute—fossilized dentine, also known as odon-
already been presented in Bennett and Mascetti tolite, ivory turquoise, bone turquoise, or French
(2003, p. 102) as turquoise jewelry. They were set turquoise. Much of this material consists of fos-
with numerous small rose-cut diamonds and a few silized mastodon ivory from Miocene-age (13–16
larger old-cut diamonds, but most prominent were a million years old) sedimentary rocks of the Gers
number of light blue to greenish blue cabochons that District between the Aquitaine and Languedoc
appeared to be turquoise. Visual examination quick- regions of southwestern France (Reiche et al., 2001).
ly revealed otherwise. Considering the historic back- The tusks are hosted by alluvial sediments (molasse
ground of these brooches, we were interested in alternating with fine sand and clay facies) that accu-
examining the blue gems in greater detail to shed mulated in basins during the erosion of the nearby
light on early turquoise imitations. Pyrenees Mountains (Crouzel, 1957; Antoine et al.,
Turquoise, a copper-bearing hydrated alu- 1997). The fossilized dentine consists mainly of fluo-
minophosphate with the chemical formula rapatite, Ca5(PO4)3F; since medieval times, local
Cistercian monks have used a heating process to
turn the material light blue (de La Brosse, 1626;
See end of article for About the Authors and Acknowledgment.
Réaumur, 1715; Fischer, 1819), which they thought
GEMS & GEMOLOGY, Vol. 47, No. 4, pp. 296–301, to be turquoise. These “stones” were originally set in medieval religious artifacts, but came into fashion
© 2011 Gemological Institute of America in the early to mid-19th century (Brown, 2007),

Figure 1. These six
brooches are set with
313 light blue stones,
the majority of which
proved to be fossilized
dentine (odontolite),
mixed with a few
turquoise and glass
cabochons. Photo by
Luc Phan, Swiss
Institute SSEF.

when fossilized dentine was recovered commercially magnification (200×) using an Olympus microscope
in southwestern France. coupled with our Renishaw Raman system. For iden-
A similar set of brooches containing odontolite tification, Raman spectra were taken on a large num-
was described by Crowningshield (1993). The pre- ber of stones, using a 514 nm argon-ion laser (Hänni
sent study offers further data on this material. Odon- et al., 1998). The spectra were collected from 1800 to
tolite is rarely encountered in the market today, 100 cm–1 Raman shift, to include the vibrational
although it is occasionally present in historic jewels range of organic compounds, such as wax and artifi-
from private collections or museums. Gemologists cial resin, used for turquoise impregnation. In a few
seldom have the opportunity to test this material in cases, spectra were collected up to 5000 cm–1 to
the laboratory. check for OH bands in the dentine.
We also conducted semiquantitative energy-dis-
MATERIALS AND METHODS persive X-ray fluorescence (EDXRF) chemical analy-
Six brooches, all of very similar style (figure 1), were sis of two cabochons, using a Thermo Fisher
investigated. Their ornamental patterns of folded Scientific Quant’X unit. These analyses, carried out
and knotted bands are characteristic of early to mid- using a series of excitation energies from 4 to 25 kV,
19th century design (Bennett and Mascetti, 2003). covered a large range of elements, from Na to those
Several French assay marks were seen on the metal with high atomic number.
mounting. In total, the brooches contained 313
opaque light blue cabochons from approximately 2 RESULTS
to 11 mm long, set with numerous small rose-cut The 313 light blue cabochons in the brooches (table
diamonds and three old-cut diamond center-stones. 1) were categorized into three groups: odontolite (288
The brooches ranged
from approximately 2.5
to 14 cm long and from TABLE 1. Gems identified in the historic “turquoise” brooches.
6.6 to 53.6 g in weight. Brooch Location in No. No. analyzed
Odontolitea Turquoise Glass
All of the pieces figure 1 cabochons by Raman
were observed micro- A Top left 94 88 87 7 0
scopically with 10–50× B Center 59 52 57 0 2
magnification. A few C Top right 57 52 52 0 5
stones were very diffi- D Bottom right 54 44 52 0 2
cult to investigate due E Bottom left 24 24 21 1 2
F Bottom center 25 24 19 2 4
to the complexity of
Total 313 284 288 10 15
the mounting. Many of
the cabochons were a Due to the mountings, a few of the odontolites could only be identified by microscopic examination;

these are also included here.

also examined at high

characteristic for elephant, mammoth, and masto-
don ivory (Campbell Pedersen, 2010).
The Raman spectra of the odontolite revealed a
distinct peak at 964 cm–1 and smaller peaks at about
1090, 580, and 430 cm–1 Raman shift (figure 5), and
only a weak, broad OH band at about 3540 cm–1.
This pattern showed a perfect correlation with fluo-
rapatite spectra taken from the SSEF reference min-
eral collection and with the published literature
(Reiche et al., 2000; Campillo et al., 2010). EDXRF
analyses of two cabochons confirmed their identity
as apatite, revealing Ca and P as major elements and
low concentrations of S, Cl, Sr, and Mn. Both analy-
ses also revealed traces of Cu.
The turquoise cabochons showed a smoothly pol-

Figure 2. Micropores were observed on the surface of

the odontolite cabochons. Photomicrograph by M. S.
Krzemnicki; magnified 30×.
• Odontolite is fossilized dentine (mastodon ivory)
stones), turquoise (10), and artificial silica glass (15). from France that has been heat treated to pro-
The odontolite cabochons all showed a micro- duce its blue coloration.
granular surface covered with a dense pattern of • This historic turquoise substitute was identified in
micropores. These very tiny pores were either round- a set of six antique brooches set with diamonds.
ed in outline (figure 2) or occurred as longitudinal • A combination of microscopic observation and
channels, depending on how they were intersected Raman spectroscopy was effective for separating
by the curved surface of the cabochon. On a macro odontolite from the turquoise and artificial silica
scale, these cabochons often showed weak banding glass also present in the brooches.
(figure 3), and in some cases a very distinct pattern of
curved intersection banding (figure 4), described as

Figure 4. Characteristic curved intersection bands

Figure 3. The odontolite displayed weak banding. were visible on several of the odontolite cabochons.
Photomicrograph by M. S. Krzemnicki; magnified 15×. Photomicrograph by M. S. Krzemnicki; magnified 20×.


582 428


644 244

1056 985 831

1600 1400 1200 1000 800 600 400 200


Figure 5. Raman spectra are shown for odontolite,

turquoise, and blue silica glass.
Figure 6. Fine brown veins are visible in this turquoise
specimen. The two neighboring cabochons are odontolite.
ished surface and even color; some also had fine Photomicrograph by M. S. Krzemnicki; magnified 15×.
irregular brown veins (figure 6). They had a slightly
more greenish blue color than the odontolite. Their
Raman spectra were characterized by a general pieces did not contain any turquoise at all.
increase in Raman signal, with a distinct doublet at Bennett and Mascetti (2003, p. 89) pictured an
~1040 cm–1 Raman shift and a series of smaller antique brooch set with diamonds and blue cabo-
peaks between 650 and 200 cm –1 , typical for chons described as odontolite and turquoise. One of
turquoise. We found no peak in the 1800–1400 cm–1 the cabochons in the photo shows a distinctly green-
range that would be expected for turquoise treated ish blue color, suggesting to the present authors that
with wax and/or stabilized with artificial resin it is turquoise, mixed with seven odontolite cabo-
(Kiefert et al., 1999). chons. We presume that mixing of these similar-
The silica glass cabochons showed a smooth sur- looking materials was common at that time. It is not
face, with some scratches and small but distinctly clear how much the jewelers actually knew about
spherical gas bubbles (figure 7). They revealed only a the materials they were using.
very weak, indistinct Raman signal characterized by
three broad bands at about 1060, 985, and 830 cm–1,
attributable to the Si-O vibrational modes of silica Figure 7. A gas bubble is apparent in this glass cabo-
glass (McMillan, 1984). chon in one of the brooches. Photomicrograph by M.
S. Krzemnicki; magnified 25×.
The brooches exemplify the fashionable use of odon-
tolite as a turquoise imitation in mid-19th century
period jewelry. This was especially true in France,
the source of the material.
Figure 8 shows the distribution of odontolite and
turquoise in the largest brooch. It contained only
seven pieces of turquoise, together with 87 odonto-
lite cabochons. The turquoise specimens were small
and rather hidden, whereas the odontolite occupied
the most prominent positions. In contrast to the
other brooches, we found no silica glass in this item.
In general, the distribution of turquoise cabochons in
the brooches seemed rather random, and three of the

Figure 8. The
largest brooch
(~14 cm long) con-
tained mostly
odontolite with
a few turquoise
Photo by M. S.

Based on its appearance and historical availabili- There was no visual indication on any of the investi-
ty, we presume that the turquoise in this jewelry gated samples of artificial blue color concentrations,
originated from classical sources in the Middle East, as would be expected for dyeing with a copper-bear-
such as Persia. They showed no indications of any ing solution (e.g., copper sulfate).
treatment (waxing, stabilization, or dyeing), as The glass imitations were uncommon in these
expected for the time period of the jewelry. brooches. Whether they were set during the crafting
With its attractive light blue color, odontolite has or during subsequent repair is not known. Similar
been used as a turquoise simulant since the Middle glass, however, has a long history as a substitute
Ages (Reiche et al., 2001). Although the heat- (Hänni et al., 1998), and is often found in fashion
induced coloration was described in the early 18th jewelry from the 19th century.
century (Réaumur, 1715; Fischer, 1819), the cause of
the blue color has been a subject of debate. Reiche et CONCLUSIONS
al. (2000, 2001) only recently showed that the oxida- What started as routine testing of a set of brooches
tion of manganese traces within the fluorapatite dur- ultimately shed light on the widespread use of a rare
ing a heating process is responsible for the blue hue turquoise imitation—odontolite—in mid-19th cen-
of the originally light gray odontolite. Using X-ray tury jewelry that was much in fashion in Western
absorption spectroscopy, these authors found that Europe. The odontolite cabochons were mixed with
heating to about 600°C under oxidizing conditions turquoise and also set with glass either at the manu-
transforms octahedrally coordinated Mn2+ into tetra- facturing stage or during subsequent repair. The
hedrally coordinated Mn5+, which substitutes for most useful approach to identifying these materials
phosphorous in the fluorapatite (Reiche et al., 2002). is a combination of microscopic observation and
The traces of Cu that we detected in the two Raman spectroscopy. Both methods are fully nonde-
odontolite cabochons using EDXRF spectroscopy structive so they can be readily applied to valuable
may result from contamination during polishing. historic objects.


Dr. Krzemnicki ( is director, Dr. Herzog is The authors thank Thomas and Ida Fä rber (Färber Collection,
analytical technician, and Dr. Zhou is a gemologist at Swiss Geneva) for loaning these brooches for our research, and for
Gemmological Institute SSEF, Basel, Switzerland. fruitful discussions about these historic jewels.

REFERENCES G&G, Vol. 19, No. 2, p. 117.
Ahmed A. (1999) Türkis aus dem ägyptischen Sinai. extraLapis Hänni H.A., Schubiger B., Kiefert L., Häberli S. (1998) Raman
No. 16: Türkise—Der Edelstein mit der Farbe des Himmels, investigations on two historical objects from Basel Cathedral:
Christian Weise Verlag, Munich, Germany, pp. 76–81. The Reliquary cross and Dorothy monstrance. G&G, Vol. 34,
Antoine P-O., Duranthon F., Tassy P. (1997) L’apport des grandes No. 2, pp. 102–113,
mammifères (Rinocérotides, Suoïdes, Proboscidiens) à la con- Hurwit K.N. (1988) Gem Trade Lab Notes: Imitation turquoise with
naissance des gisements du Miocène d’Aquitaine (France). In J.- “veins” and pyrite. G&G, Vol. 24, No. 1, p. 52.
P. Aguilar, S. Legendre., and J. Michaux, Eds., Actes du Congrès Kane R.E. (1985) Gem Trade Lab Notes: Turquoise simulant, dyed
BiochoM’97, Mémoires Travaux E.P.H.E., Institut Montpeillier, magnesite. G&G, Vol. 21, No. 1, pp. 47–48.
Vol. 21, pp. 581–591. Kiefert L., Hänni H.A., Chalain J.-P., Weber W. (1999)
Arnould H., Poirot J-.P. (1975) Infra-red reflection spectra of Identification of filler substances in emeralds by infrared and
turquoise (natural and synthetic) and its substitutes. Journal of Raman spectroscopy. Journal of Gemmology, Vol. 26, No. 8,
Gemmology, Vol. 14, pp. 375–377. pp. 501–520.
Bennett D., Mascetti D. (2003) Understanding Jewellery. Antique Lind T., Schmetzer K., Bank H. (1983) The identification of
Collectors Club Ltd., Woodbridge, Suffolk, England, 494 pp. turquoise by infrared spectroscopy and X-ray powder diffrac-
Brown G. (2007) Rare ivories—Challenging identifications. tion. G&G, Vol. 19, No. 3, pp. 164–168,
Lecture presented at the Federal Conference of the Gemmo- 10.5741/GEMS.19.3.164.
logical Association of Australia, Hobart, Tasmania, May 19, McMillan P. (1984) Structural studies of silicate glasses and melts—Applications and limitations of Raman spectroscopy.
Campbell Pedersen M. (2010) Gem and Ornamental Materials of American Mineralogist, Vol. 69, pp. 622–644.
Organic Origin. NAG Press, London. Réaumur R. (1715) Observations sur les mines de turquoises du
Campillo M., Lacharmoise P.D., Reparaz J.S., Goni A.R., Valiente royaume; sur la nature de la matière qu’on y trouve, et sur la
M. (2010) On the assessment of hydroxyapatite fluoridation matière dont on lui donne la couleur. Memoires de l’Académie
by means of Raman scattering. Journal of Chemical Physics, Royale des Sciences, pp. 174–202.
Vol. 132, No. 24, article no. 244501 [5 pp.], Reiche I., Vignaud C., Menu M. (2000) Heat induced transforma-
10.1063/1.3428556. tion of fossil mastodon ivory into turquoise ‘odontolite’: Struc-
Chalker K., Dubin L.S., Whiteley P.M. (2004) Totems to tural and elemental characterisation. Solid State Sciences, Vol.
Turquoise: Native North American Jewelry Arts of the North- 2, No. 6, pp. 625–636,
west and Southwest. Published in association with the 2558(00)01067-0.
American Museum of Natural History by Harry N. Abrams, Reiche I., Vignaud C., Champagnon B., Panczer G., Brouder C.,
New York, 224 pp. Morin G., Solé V.A., Charlet L., Menu M. (2001) From
Crouzel F. (1957) Le Miocène continental du Bassin d’Aquitaine. mastodon ivory to gemstone: The origin of turquoise color in
Bulletin du Service de la Carte Géologique de la France, Vol. odontolite. American Mineralogist, Vol. 86, pp. 1519–1524.
54, No. 248, 264 pp. Reiche I., Morin G., Brouder C., Solé V.A., Petit P.-E., Vignaudi C.,
Crowningshield G.R. (1993) Gem Trade Lab Notes: Odontolite. Calligaro T., Menu M. (2002) Manganese accommodation in
G&G, Vol. 29, No. 2, p. 127. fossilised mastodon ivory and heat-induced colour transforma-
de La Brosse G. (1626) Livre sur la Nature, vertu et Utilité des tion: Evidence by EXAFS. European Journal of Mineralogy,
Plantes. Bibliothèque Interuniversitaire de Médecine et d’Odon- Vol. 14, pp. 1069–1073,
tologie, Paris. 1221/2002/0014-10693.
Fischer G. (1819) Essay on the turquoise and the calcite. In T. Salanne C. (2009) Etude de la Turquoise, de ses Traitements et
Thomson, Ed., Annals of Philosophy, Vol. 14, pp. 406–420. Imitations. Diplôme d’Université de Gemmologie, University
Fryer C.W. (1983) Gem Trade Lab Notes: Turquoise imitation. of Nantes, France, 88 pp.

Rapid Communications



Jian Zhang, Taijin Lu, Manjun Wang, and Hua Chen

um blue color, called “London Blue.” Unfortunately, this

A systematic study of 15 neutron-irradiated treatment also produces radioactivity from the nuclides of
blue topaz samples was conducted using trace-element impurities in topaz (Crowningshield, 1981;
Ashbaugh, 1988), such as Fe, Mn, Co, Zn, Sb, Ta, Cs, Sc,
high-purity germanium (HPGe) digital
and Tb (Foord et al., 1988; Northrup and Reeder, 1994).
gamma-ray spectroscopy. The specific These radioactive nuclides, which have various half-lives
activity of the detected radionuclides (134Cs, (table 1), may emit γ-rays and beta (β) particles of varying
182Ta, 46Sc, and/or 160Tb) was measured,
radiation intensity (Ashbaugh, 1991; Nelson, 1991). While
and the decay pattern of the irradiated a high dose of γ-rays and β-particles poses danger, a very
topaz was determined. Based on the time low dose is not harmful. Therefore, blue topaz colored by
elapsed since their removal from the nucle- neutron irradiation requires a quarantine period to allow
ar reactor, the amount of time required for the residual radioactivity to reach a safe level (referred to,
the residual radioactivity to decay to safe e.g., as the exemption level) of less than 74 becquerels per
levels was calculated. Most of the samples gram (Bq/g) or 2 nanocuries per gram (nCi/g; Ashbaugh,
were safe at the time of the first measure- 1988).
Investigations of blue topaz have focused mainly on
ment (95 days after irradiation), but higher
irradiation methods and the cause of the coloration, with
concentrations of radionuclide impurities in less emphasis on the detection of residual radioactivity.
some samples will require them to be quar- Ashbaugh (1988) discussed the radioactivity of colored
antined for several years. stones treated by irradiation, as well as potential health
hazards and government regulations. Based on these find-
ings, GIA offered testing services for irradiated gems from
1991 through 2006. However, that article did not address
blue topaz treated by neutron irradiation. Guo et al. (2000)

R adioactivity is one of the most discussed topics sur-

rounding blue topaz (figure 1), which is commonly
treated by irradiation from near-colorless starting
material. To maintain consumer confidence, it is impor-
tant to ensure that these gems do not contain dangerous
measured the specific activity of the total alpha (α) and β
radioactivity of crushed blue topaz after irradiation using
an FJ-2603 low-level α- and β-radiation measuring device
and a BH1216A low-background measuring instrument.
While that effort was helpful in exploring the radioactive
levels of residual radioactivity. Three major irradiation decay of irradiated blue topaz, the specific nuclides
methods are used: gamma (γ), neutron, and electron-beam involved were not determined. Helal et al. (2006) analyzed
irradiation (Nassau, 1985; Ashbaugh, 1988). Of these, neu- the trace elements in topaz samples before and after neu-
tron irradiation creates perhaps the most beautiful medi- tron irradiation using ICP-MS and a high-purity germani-
um (HPGe) detector. This study proved that radioactivity
in blue topaz colored by neutron irradiation is related to
See end of article for About the Authors and Acknowledgments. variations in the trace elements in the topaz. However, the
GEMS & GEMOLOGY, Vol. 47, No. 4, pp. 302–307, decay pattern of the radionuclides was unclear. This arti- cle studies the residual radioactivity in blue topaz treated
© 2011 Gemological Institute of America by neutron irradiation, identifies the specific radionuclides


Figure 1. Most blue
topaz on the market is
irradiated to enhance
its color. These stones
(~4–10 ct) were treated
by neutron irradiation.
Photo by J. Zhang.

sis was performed on all 15 samples (after irradiation) and

TABLE 1. Half-lives of radioactive nuclides in
on one additional untreated sample from the same locality,
irradiated blue topaz.
using an ARL Quant'X spectrometer at the National
Radionuclide Half-life (days) Gemstone Testing Center in Beijing.
59Fe 44.51
The important color centers in topaz are produced after
124Sb 60.20 12 hours of irradiation at 1.2 × 1019 neutrons/(cm2 × sec).
160Tb 72.30 These conditions were used in this study to produce the
46Sc 83.81 colors shown in figure 2. The irradiation was conducted in
182Ta 114.4 a “light water” (ordinary water) nuclear reactor at the
65Zn 243.8 China Institute of Atomic Energy in Beijing. The samples
54Mn 312.2 were placed in cadmium-lined containers to reduce the
134Cs 753.7 amount of the thermal neutrons caused by absorption, and
60Co 1,924 also to increase the amount of the fast, color-producing
A GEM-30185-Plus Despec HPGe digital gamma-ray
spectrometer (figure 3) was used to measure the residual
involved, and calculates quarantine periods for the most
radioactivity of the irradiated blue topaz. The Ge semicon-
highly radioactive samples to reach the exemption level.
ductor detector had a relative efficiency of 25%, an energy
resolution of 1.96 keV at 1332.5 keV for 60Co, and a peak
MATERIALS AND METHODS ratio of 48. The values measured represent the combined
Fifteen samples of cut but unpolished near-colorless topaz radioactivity of specific nuclides in the process of decay.
(weighing 0.75 to ~2.0 g) of pegmatite origin from China’s The measurements were semiquantitative, obtained for
Guangdong Province were studied. EDXRF chemical analy- 40,000 seconds of active time.

Figure 2. These irradiated

topaz samples (weighing
0.75 to ~2.0 g) were used in
this study. From left to right,
the top row shows samples
numbered Topaz-1 to Topaz-
7, and the bottom row
shows Topaz-8 to Topaz-15.
Photo by J. Zhang.


EDXRF spectroscopy showed traces of Mg, Ca, Na, K, and
Cl in all samples, while Fe, Cs, Sr, Sc, Ta, and Tb were
detected in some of them. Gamma-ray spectroscopy
revealed the variable presence of four radionuclides in the
irradiated topaz: 134Cs, 182Ta, 46Sc, and/or 160Tb (table 2).
When the samples were first tested 95 days after irradia-
tion, four of them showed residual radioactivity above the
exemption level. The residual radioactivity of the other 11
samples had decayed below the exemption level or the
detection limit of the spectrometer.

Figure 3. The HPGe digital spectrometer at the China • Neutron irradiation is commonly used to create
Institute of Atomic Energy in Beijing was used in this an attractive “London Blue” color in near-color-
study. Photo by J. Zhang. less topaz.
• Topaz may contain trace impurities that become
radioactive after neutron irradiation.
To explore the decay pattern of the irradiated blue • Gamma-ray spectroscopy showed that irradiated
topaz, we conducted four measurements of the residual topaz samples from China contained up to four
radioactivity in the 15 samples. The samples were radionuclides.
removed from the reactor on September 28, 2005, and • Most of the samples were safe to handle when
gamma-ray spectroscopy was subsequently performed on measured 95 days after irradiation, but some will
January 1, 2006 (day 95); March 14 (day 167); April 19 require several years to “cool down.”
(day 203); and November 30 (day 428). We tested all sam-
ples at day 95, and then focused on four samples (figure 4)
with residual radioactivity higher than the exemption
value (74 Bq/g). The goal of studying the decay pattern of irradiated
blue topaz is to determine the time needed for the residual
radioactivity to decay to a safe level (Miraglia, 1986;
Miraglia and Cunningham, 1988). The half-life decay for-
TABLE 2. Residual radioactivity (Bq/g) of irradiated mula of radioactive nuclides is necessary for these calcula-
blue topaz samples after four periods of decay.a tions. De Soete et al. (1972) defined the formula as follows:
If the probability of decay for radioactive elements in unit
Decay time (days)
time is 1/τ, and if the number of radioactive elements is N,
no. 95 167 203 428 then the number of decayed radioactive elements in the
time span of dt should be dN, where
Topaz-8 Cs 417.6 392.1 379.2 309.3
Ta 19,800 12,800 10,290 2,668 dN/N = −1/τ dt (1)
Total 20,218 13,192 10,669 2,977 In formula (1), the negative sign indicates a reduction
Topaz-9 134
Cs 1,617 1,512 1,462 1,187 in the number of subatomic particles that constitute ioniz-
Sc 296.3 162.1 120.7 18.4 ing radiation. Integrating both sides of the equation yields:
Ta 495.9 321.1 258.1 66.8
N = N0e−t/τ (2)
Total 2,409 1,996 1,840 1,273
In formula (2), N0 is the original value of radioactivity
Topaz-10 Ta 195,000 125,800 100,800 25,750 from a given nuclide, and τ represents the decay time con-
Total 195,000 125,800 100,800 25,750 stant, after which the element has been reduced to e−1 of
134 the original value of radioactivity. The relationship
Topaz-15 Cs 106.2 100.2 96.5 78.7
Sc 237.9 132.1 98.6 15.0 between τ and half-life (T1/2) can be expressed as:
Tb 3,909 1,961 1,389 160.5 T1/2 = 0.693τ (3)
Total 4,253 2,193 1,585 254.2 Applying formula (3) to formula (2), we can derive the
Although Fe was detected in the samples by EDXRF analysis, no half-life decay formula of radioactive nuclides:
iron radionuclides were detected by gamma-ray spectroscopy.
N = N0e−0.693t/T /
1 2


calculated. Most importantly, the time required for the
residual radioactivity to decay below the exemption level
(74 Bq/g) can be derived using the formula; these times are
shown in table 4 for the four samples.

Blue topaz (e.g., figure 5) typically takes two to three years
to decay below the exemption level after neutron irradia-
tion. The actual quarantine time necessary for specific
samples depends on several factors. There are three scenar-
Figure 4. These four samples (9.5–16.5 mm long) had ios that can contribute to high levels of residual radioactiv-
residual radioactivity higher than the exemption ity in irradiated gems.
level when initially measured after 95 days. From left The first is the presence of activated impurities with a
to right: Topaz-8, Topaz-9, Topaz-10, and Topaz-15. long half-life (i.e., a comparatively slow rate of decay). For
Photo by J. Zhang. example, 134Cs has a half-life of 2.06 years. This radionu-
clide was present in three of the four topaz samples show-
ing residual radioactivity, and for Topaz-9 the time required
In formula (4), N0 is the initial value of radioactivity to decay below the exemption level was calculated at 3,440
and N is the unknown value of radioactivity, expressed in days (~9.4 years).
units of Bq. T1/2 is the half-life of the nuclide, and t is its Second, high concentrations of activated impurities
decay time. (even those that do not have a long half-life) may produce
The specific activity of each radionuclide present in a high residual radioactivity. Sample Topaz-10, for example,
given sample was calculated. The values obtained at the contained a large amount of 182Ta (half-life of 114 days).
first stage of radiation detection (after 95 days) can be set Consequently, the sample’s radioactivity upon removal
as the initial values. Values for each subsequent stage can from the reactor was 347,600 Bq/g, and it took 1,391 days
then be calculated and compared with the detected values, (~3.8 years) for the sample to decay below the exemption
as shown in table 3. In addition, the radioactivity of the level.
samples upon their removal from the reactor (day 0) can be The third scenario contributing to high levels of residual

TABLE 3. Detected and calculated specific activity (Bq/g) of radionuclides in

irradiated blue topaz.

Decay time (days)

Sample no. Nuclide Method
0 95 167 203 428
Topaz-8 Cs Detected 417.6 392.1 379.2 309.3
Calculated 455.8 390.8 378.0 307.2
Ta Detected 19,800 12,800 10,290 2,668
Calculated 35,280 12,780 10,270 2,664
Topaz-9 134Cs Detected 1,617 1,512 1,462 1,187
Calculated 1,764 1,513 1,464 1,190
46Sc Detected 296.3 162.1 120.7 18.4
Calculated 650.0 163.4 121.3 18.9
182Ta Detected 495.9 321.1 258.1 66.8
Calculated 877.7 320.1 257.2 66.5
Topaz-10 182Ta Detected 195,000 125,800 100,800 25,750
Calculated 347,600 125,900 101,100 25,770
Topaz-15 134Cs Detected 106.2 100.2 96.5 78.7
Calculated 115.7 99.2 96.0 78.0
46Sc Detected 237.9 132.1 98.6 15.0
Calculated 521.8 131.1 97.4 15.1
160Tb Detected 3,909 1,961 1,389 160.5
Calculated 9,717 1,960 1,388 160.6


TABLE 4. Decay time needed for irradiated blue topaz samples to reach
the exemption level of 74 Bq/g.

Decay time (days)

Sample no. Nuclide
0 95 428 703 1,391 1,975 3,440
Topaz-8 134Cs 455.8 417.6 309.3 238.4 139.8 73.8
Ta 35,280 19,800 2,668 491.5 14.5 0.2
Total 35,736 20,218 2,997.3 729.9 154.3 74.0
Topaz-9 Cs 1,764 1,617 1,187.3 922.5 540.7 285.6 74.0
46Sc 650.0 296.3 18.4 1.9 0 0 0
182Ta 877.7 495.9 66.8 12.2 0.4 0 0
Total 3,291 2,409 1,272 936.6 541.1 285.6 74.0
Topaz-10 Ta 347,600 195,000 25,750 4,843 73.9
Total 347,600 195,000 25,750 4,843 73.9
Topaz-15 134Cs 115.7 106.2 78.7 60.5
Sc 521.8 237.9 15.0 1.6
160Tb 9,717 3,909 160.5 11.5
Total 10,350 3,434 254.2 73.6

radioactivity is the presence of activated

impurities with a long half-life combined
with relatively high concentrations.
The Chinese topaz studied for this article
should be assumed to contain different
trace-element impurities than starting
material from other sources (e.g., Brazil), and
therefore the decay times presented in this
study are not representative of all neutron-
irradiated topaz on the market. Still, the
procedure and data presented here provide
useful information for evaluating the residu-
al radioactivity in topaz treated by neutron
irradiation, regardless of locality.

Figure 5. Treated blue topaz (here,

85.79–243.66 ct) is commonly encoun-
tered in the global gem market, and
neutron-irradiated samples must
undergo appropriate safeguards to
ensure that they do not contain danger-
ous levels of residual radioactivity.
Photo by Robert Weldon; from top to
bottom, GIA Collection nos. 16203,
30886, 31942, 30889, and 31947.


Mr. Zhang ( is an engineer, Dr. Lu is The authors thank Ms. Xiuqing Gao for irradiating
chief researcher, Ms. Wang is a professor, and Ms. the topaz samples, and Mr. Yongbao Gao for per-
Chen is director of the research department, at the forming the gamma-ray spectroscopy; both are
National Gems & Jewellery Technology Administrative located at the China Institute of Atomic Energy in
Center, in Beijing, China. Beijing.

REFERENCES Irradiated topaz in the reactor. VIII Radiation Physics &

Ashbaugh C.E. (1988) Gemstone irradiation and radioactivity. Protection Conference, Beni Suef - Fayoum, Egypt, November
G&G, Vol. 24, No. 4, pp. 196–213, 13–15, pp. 447–451.
GEMS.24.4.196. Miraglia F.J. (1986) Distribution of Products Irradiated in Research
Ashbaugh C.E. III (1991) Radioactive and radiation treated gem- Reactors. U.S. Nuclear Regulatory Commission Generic
stones. Radioactivity & Radiochemistry, Vol. 2, No. 1, pp. Letter 86-11, June 25.
42–57. Miraglia F.J., Cunningham R.E. (1988) Distribution of Gems
Crowningshield R. (1981) Irradiated topaz and radioactivity. Irradiated in Research Reactors. U.S. Nuclear Regulatory
G&G, Vol. 17, No. 4, pp. 215–217, Commission Generic Letter 88-04, February 23.
GEMS.17.4.215. Nassau K. (1985) Altering the color of topaz. G&G, Vol. 21, No. 1,
Foord E.E., Jackson L.L., Taggart J.E., Crock J.G., King T.V.V. pp. 26–34,
(1988) Topaz: Environments of crystallization, crystal chem- Nelson K.L. (1991) Health Risk Assessment of Irradiated Topaz.
istry, and infrared spectra. Mineralogical Record, Vol. 26, pp. Unpublished PhD thesis, University of Minnesota.
69–70. Northrup P.A., Reeder R.J. (1994) Evidence for the importance of
Guo Y.C., Liu X.L., Huang J.Z. (2000) Radioactivity and decay pat- growth-surface structure to trace element incorporation in
tern of topaz colored by neutron radiation. China Occu- topaz. American Mineralogist, Vol. 79, pp. 1167–1175.
pational Medicine, Vol. 27, No. 2, pp. 19–21 [in Chinese]. de Soete D., Gijbels R., Hoste J. (1972) Neutron Activation Analy-
Helal A.I., Zahran N.F., Gomaa M.A.M., Salama S. (2006) sis. John Wiley & Sons, New York.

Dr. Ilaria Adamo ❁ Dr. Vladimir

Balitsky ❁ Dudley Blauwet ❁ Bill
Boyajian ❁ Dr. Christopher M.
Breeding ❁ Maggie Campbell-Pedersen
❁ Dr. Francois Farges ❁ Dr. David
Fisher ❁ Dr. Harald Gabasch ❁ Dr.
GEMS & GEMOLOGY requires that all articles Eloïse Gaillou ❁ Rui Galopim de
undergo a peer review process in which each Carvalho ❁ Al Gilbertson ❁ Dr.
Gaston Giuliani ❁ Dr. Lee Groat ❁
manuscript is evaluated by at least three
Dr. John Gurney ❁ Thomas Hain-
experts in the field. This process is vital to the schwang ❁ Hertz Hasenfeld ❁
accuracy and readability of the published article. Richard Hughes ❁ Dr. Stefanos
Because members of our Editorial Review Karampelas ❁ John King ❁ William
Board cannot have expertise in every area, we Larson ❁ Elise Misiorowski ❁
sometimes call on other experts to share their Vincent Pardieu ❁ Dr. Adolf Peretti ❁
Dr. Federico Pezzotta ❁ John Pollard
intellect and insight. In addition to the members
❁ Dr. Ilene Reinitz ❁ Dr. Benjamin
of our Editorial Review Board, we extend a Rondeau ❁ Dr. Andy Shen ❁ Russell
heartfelt thanks to the following individuals who Shor ❁ Elisabeth Strack ❁ Carol
reviewed manuscripts for G&G in 2010–2011. Stockton ❁ Ursula Wehrmeister


Thomas M. Moses Shane F. McClure | GIA Laboratory

color. FTIR spectroscopy showed the

diamond was a type IaA with a very
A Rare Fancy Vivid Purple Diamond high concentration of aggregated
In GIA’s colored diamond grading sys- nitrogen, predominantly in the A
tem, predominantly purple diamonds form. We also noted a very weak
can occur in the red-purple, reddish absorption at 3107 cm−1, attributed to
purple, and purple hue ranges. While hydrogen.
none of these are common, diamonds The outstanding feature of this
in the unmodified purple range are diamond was its depth of color, which
the rarest. placed it in the Fancy Vivid range. On
GIA first encountered purple dia- the rare occasions the lab has seen
monds in the 1970s. Their color is due purple diamonds, they typically have
to plastic deformation of the crystal a dark or unsaturated color, resulting
structure that occurred during their in color grades modified by “gray” or
geologic history, resulting in parallel “grayish.” It is unusual for GIA to
glide planes (referred to as graining). encounter more than a handful of pur-
This graining exhibits pink and ple diamonds, regardless of their
brown dichroism, observed under depth of color, in any given year. The
magnification (see S. Titkov et al., rarity is exponentially greater for a
“Natural-color purple diamonds from Figure 1. This extremely rare 0.81 diamond with a very strong purple
Siberia,” Spring 2008 G&G, pp. ct diamond was color graded hue.
56–63). It is extremely rare to see a Fancy Vivid purple. Jason Darley, Paul Johnson, and
pure purple hue resulting from such John King
plastic deformation.
The New York laboratory recently Figure 2. The purple coloration
had the opportunity to examine a 0.81 ible through the pavilion (figure 2). in the diamond was concentrat-
ct purple diamond with unusually These planes showed a strong concen- ed along parallel glide planes.
strong color (figure 1). The cut-cor- tration of purple color, while the sur- Magnified 60×.
nered rectangular modified brilliant rounding area appeared nearly color-
was faceted so the graining reflected less. The stone fluoresced weak yel-
the purple color throughout the stone, low to long-wave UV radiation and
maximizing its intensity. With mag- was inert to short-wave UV. Under
nification and diffused lighting, well- the very strong short-wave UV of the
defined parallel glide planes were vis- DiamondView, the glide planes dis-
played clear green fluorescence,
which is attributed to the localized
Editors’ note: All items were written by staff distribution of the H3 optical center.
members of the GIA Laboratory. The H3 center was also present in the
UV-Vis absorption spectrum taken at
GEMS & GEMOLOGY, Vol. 47, No. 4, pp. 308–315,
liquid-nitrogen temperature, along
with a strong absorption band cen-
© 2011 Gemological Institute of America tered at ~550 nm, which clearly con-
tributed to the predominantly purple


tion of hydrogen produced such a dark
appearance that the diamond received
a Fancy black color grade.
We compared this heart shape to a
0.20 ct modified pear shape (figure 3,
right) with similar overall spectro-
scopic properties, also type Ia, that
received a Fancy grayish blue color
grade. In this diamond, however, the
hydrogen impurity did not cause such
a strong absorption. The difference
between the two stones was also
clearly displayed in their Vis-NIR
absorption spectra in the 400–900 nm
range. Both showed typical absorp-
Figure 3. These two diamonds, a 0.85 ct Fancy black heart shape (left) tions for gray to blue to violet dia-
and a 0.20 ct Fancy grayish blue modified pear shape (right), had similar monds, such as bands near 500 and
hydrogen-related spectroscopic properties but dramatically different color. 700 nm, attributed to hydrogen. The
much stronger absorption in the heart
shape blocked most of the light
return, making it appear virtually
A Strongly Purple-Colored Australia,” Spring 2009 G&G, pp. black. Its spectrum was very noisy
Black Diamond 20–37). The New York lab recently because of the absorbance, and the
tested such a diamond with an unusu- spectrometer detected only a very
Hydrogen-rich diamonds with gray to ally dark tone. weak signal (figure 4). Only with mag-
blue to violet colors are rare, but they The 0.85 ct type Ia heart shape nification and fiber-optic illumina-
are well documented in the literature (figure 3, left) showed extremely tion could we observe fine clouds
(see, e.g., C. H. van der Bogert et al., strong hydrogen-related features (e.g., associated with hydrogen in such dia-
“Gray-to-blue-to-violet hydrogen-rich 3107 cm−1) in its mid-infrared absorp- monds (figure 5); the stone’s true deep
diamonds from the Argyle mine, tion spectrum. The high concentra- purple bodycolor was also seen.
This rare diamond provides a good
example of an extreme end of the
color range of hydrogen-rich dia-
Figure 4. The Vis-NIR absorption spectrum of the Fancy black heart shape monds, in which the purple color was
showed a weak signal and significant noise. so deep that the stone appeared black.
Paul Johnson


Figure 5. With magnification and

fiber-optic illumination, the
heart-shaped diamond’s deep pur-
ple bodycolor became apparent.

400 500 600 700 800 900



Figure 6. In the DiamondView, this 0.99 ct HPHT-treated diamond (left)
shows four-fold symmetry that appears similar to the cross pattern typi- Figure 7. The HPHT-treated dia-
cally encountered in HPHT synthetics (right, 1.00 ct). mond’s mottled strain pattern
indicated natural origin. Image
width: 4.35 mm.
HPHT-Treated Diamond with the sample had been subjected to HPHT
Fluorescence Pattern of an conditions, therefore suggesting it
HPHT-Grown Synthetic was an HPHT-grown synthetic.
Nevertheless, microscopic examina- to its altered morphology. The fact
Synthetic diamonds grown at high- tion with crossed polarizers revealed that this diamond was chosen for
pressure, high-temperature (HPHT) a mottled strain pattern typical of HPHT treatment was an interesting
conditions are no longer rare in the natural type IIa diamonds (figure 7), coincidence.
market. One method to distinguish and additional peaks in the PL spec- Sally Eaton-Magaña
them from natural diamonds is imag- tra confirmed it was an HPHT-treat-
ing their growth patterns with ultra- ed natural diamond.
violet-activated fluorescence, such as DiamondView imaging can be
with the DiamondView instrument. diagnostic for synthetic diamonds,
Type IIb Diamond with Long
The growth structure of HPHT particularly those of HPHT origin. Phosphorescence
synthetics is typically quite distinc- But careful consideration of both It is well known that type II synthetic
tive and, in most cases, eloquently DiamondView images and spectral diamonds and some natural diamonds
displayed in their fluorescence data (particularly PL) is usually neces- (including type IIb) are phosphores-
images. Because HPHT synthetic dia- sary to definitively identify natural or cent. In natural diamonds the effect
monds form under different condi- synthetic origin. The cross-shaped usually lasts for only a few seconds.
tions than their natural counterparts, growth pattern in the present dia- However, a 4.23 ct emerald cut (figure
the growth rates of their various crys- mond may be due to formation under 8) recently submitted to the Carlsbad
tal faces are generally quite different. different conditions (i.e., lower tem- laboratory proved to be an exception.
The resulting morphology of HPHT peratures) than those typically experi- Identified as an untreated type IIb dia-
products is typically cubo-octahedral, enced by natural diamonds, which led mond with D color and IF clarity, it
whereas natural diamonds grow as
octahedra. The overwhelming majori-
ty of HPHT synthetics show a diag- Figure 8. This 4.23 ct type IIb diamond is shown before (left) and after
nostic cross shape in their lumines- (right) exposure to short-wave UV radiation for 10 seconds. The photo on
cence pattern. Sometimes, however, the right shows the diamond’s strong blue phosphorescence.
natural diamonds exhibit four-fold
symmetry that may be confused with
this cross pattern.
One such instance was seen
recently in the Carlsbad laboratory. A
0.99 ct F-color type IIa round brilliant
underwent advanced testing to deter-
mine if it was natural, treated, or syn-
thetic. The stone’s DiamondView
image (figure 6, left) showed a config-
uration resembling the cross pattern
(figure 6, right), and its photolumi-
nescence (PL) spectra indicated the


and other blue diamonds,” Geology,
Vol. 36, No. 1, 2008, pp. 83–86).
PHOSPHORESCENCE SPECTRUM Figure 10 shows the relationship
8000 between apparent bodycolor and
phosphorescence half-life in 300 natu-
ral type IIb diamonds examined in the

INTENSITY (counts)
GIA laboratory. The present dia-
mond’s 500 nm band had a much
4000 longer half-life and a stronger intensi-
ty than most natural type IIb dia-
monds; this combination produced
the unusually long phosphorescence.
This diamond also exhibited ther-
0 moluminescence, as described in the
Spring 2011 Lab Notes (pp. 50–51). It
was immersed in a liquid-nitrogen

bath (–196ºC) and simultaneously

60 exposed to short-wave UV. After

400 500 600 700 removal from the bath, it warmed up
WAVELENGTH (nm) rapidly and displayed a flash of blue
thermoluminescence. After a few sec-
onds of additional warming, it exhib-
Figure 9. The three-dimensional spectral plot for the diamond in figure 8 ited the long-lasting phosphorescence
shows its long-lasting phosphorescence corresponding to the 500 nm band. described above. A second video doc-
The diamond was exposed to UV radiation for 20 seconds and data were umenting both luminescence behav-
collected over one second intervals. iors is available in the G&G Data
Andy H. Shen and
Sally Eaton-Magaña
displayed very weak to weak greenish Magaña et al., “Using phosphores-
blue fluorescence to long- and short- cence as a fingerprint for the Hope
wave UV radiation. During UV expo-
sure, the fluorescence appeared to
intensify due to the sample’s strong Figure 10. Data compiled for 300 natural type IIb diamonds show varia-
phosphorescence. After 5 seconds of tions in the average half-life of phosphorescence bands (with error bars
exposure to short-wave UV radiation, showing standard deviations) at 500 and 660 nm according to their body-
the diamond displayed bright blue color. Colorless type IIb diamonds generally do not show a red phospho-
phosphorescence (see video in the rescence band at 660 nm, and their blue 500 nm band has a long half-life.
G&G Data Depository at Type IIb diamonds with stronger blue coloration have both 500 and 660
gandg), and it continued to luminesce nm bands, and the half-life of their blue phosphorescence is much shorter.
for several minutes.
Phosphorescence spectroscopy
revealed a band at 500 nm (also docu-
mented in type IIb diamonds by S. 18
Average 500 nm band half-life
Eaton-Magaña et al., “Luminescence

16 Average 660 nm band half-life

of the Hope diamond and other blue 14

diamonds,” Fall 2006 G&G, pp.
95–96) that was responsible for the
blue phosphorescence. Figure 9 shows 10

the spectra collected over a period of 8

60 seconds in a three-dimensional 6
plot. At the end of 60 seconds, the 500
nm phosphorescence band was still
visible. The calculated half-life of this 2

band was 8.4 seconds. By comparison, 0

the Hope diamond’s 500 nm band D-Z FANCY LIGHT BLUE FANCY DEEP
half-life is only 1.8 seconds (S. Eaton-


natural opal. The refractive index was deposit, this seems likely given its
1.45, but we did not test specific grav- appearance and the high volume of
ity because of the presence of the material recently mined there.
“matrix.” The sample was inert to Although this opal did not show
long-wave UV radiation and gave a hydrophane character, in one of
weak greenish yellow reaction to these contributors’ (NR) experience,
short-wave UV. Notably, the fractures not all Wollo opal is hydrophane
fluoresced weak yellow to both long- type. In addition, some of the orangy
and short-wave UV. yellow Wollo opal seems to be prone
Microscopic examination revealed to crazing.
a somewhat columnar structure and As with any new source, artificial-
bodycolor similar to material from ly enhanced lesser-quality material
Wollo Province, Ethiopia. A dense net- will eventually make its way to the
work of very low-relief fractures was market as treatments are developed to
also seen throughout the stone; their create salable products. Even though
subtle appearance immediately sug- very high-quality untreated Ethiopian
gested clarity enhancement. Closer opal is readily available, buyers should
Figure 11. This 67.17 ct opal was
examination showed flattened gas be aware that an increasing amount of
clarity enhanced with some type
bubbles in many of the fractures from treated material is appearing in the
of oil.
incomplete filling (figure 12, left). trade, including clarity-enhanced sam-
When exposed to a relatively low tem- ples such as this one and others sub-
OPAL perature thermal probe, the fractures jected to forms of color modification
began to sweat (figure 12, right). (see, e.g., N. Renfro and S. F. McClure,
Clarity-Enhanced, with Because of the mobile nature of the “A new dyed purple opal,” www.gia.
Artificial Matrix clarity enhancing substance, we con- edu/research-resources/news-from-
The Carlsbad laboratory recently cluded that the stone was treated with research). As with all treatments,
received a 67.17 ct opal cabochon (fig- some variety of oil rather than an these are acceptable as long as proper
ure 11) for identification. The stone epoxy resin. disclosure is given to the buyer.
showed an orangy yellow bodycolor The “matrix” on the back of the Nathan Renfro and Phil York
and moderate play-of-color. It also had cabochon consisted of a hard gray
an unusual mottled gray “matrix” on resin-like material that showed swirl
the back and felt slightly greasy to the marks and also some grinding marks,
Ethiopian Black Opal
touch. In reflected light, a network of indicating it had been applied during
fine lines on the surface revealed a or after the lapidary process. This The Carlsbad laboratory recently
large amount of crazing, but these layer added significant weight and examined a 10.67 ct black opal (figure
lines were not easily seen upon initial likely helped hold the highly frac- 13), reportedly from the Wollo
examination. tured stone together. Province of Ethiopia. Because black is
Gemological examination pro- While we do not know for certain not a common bodycolor for opal
duced measurements consistent with if this stone was from the Wollo from this important locality, and
there are increasing reports of this
material being treated (e.g., the article
Figure 12. Low-relief fractures were seen throughout the opal, many of by N. Renfro and S. F. McClure in this
which showed flattened gas bubbles near the surface (left). Droplets of issue, pp. 260–270), we sought to
oil readily sweated out of the fractures when exposed to low temperature determine the cause of color in this
heat from a thermal probe (right, reflected light). Magnified 40×. sample. Also submitted was an
approximately 1 kg parcel of opal nod-
ules, on which the client permitted
destructive testing to investigate the
cause of the bodycolor.
Standard gemological properties
from the cabochon were consistent
with those expected for opal, includ-
ing an RI of 1.45 and a hydrostatic SG
of 2.05. It was inert to long- and short-
wave UV radiation. Microscopic
examination revealed small brown
dendritic inclusions, a “digit-pattern”


columnar structure, and nonphenom- While treated black Ethiopian
enal potch areas with a greenish cast. material has been documented (e.g.,
These microscopic observations and
physical properties are consistent TreatmentinWolloOpal.pdf), opal en-
with features seen in other colors of thusiasts will certainly appreciate
Wollo opal. these natural-color gems from the
The rough material had the same Wollo Province.
general appearance and gemological Nathan Renfro
properties as the cabochon, except
that the nodules displayed color zon-
ing. One of the rough pieces was
sliced in half, revealing a dark brown Coated Bead-Cultured
to black core with a light brown to Freshwater PEARLS
near-colorless perimeter (figure 14). In September 2011, a strand of unusu-
This structure suggested natural Figure 14. This 8.37 ct slice from ally large white baroque “pearls” was
color, as the reverse pattern (dark a black opal nodule shows a dark submitted to the New York laborato-
perimeter with light core) would be core with a near-colorless to light ry for identification. The 17 pieces
expected if the dark color had been brown perimeter, suggestive of measured 22.43 × 17.60 × 14.17 mm
artificially introduced. natural color. LA-ICP-MS analy- to 29.46 × 19.64 × 16.50 mm. They
We performed LA-ICP-MS chemi- sis of the core revealed high con- had a noticeably unnatural color and
cal analysis of one slice, and Mn was centrations of Mn, which is surface appearance, yet their baroque
the only element that showed distinct responsible for the bodycolor. shape was typical of some freshwater
variations between the dark core cultured pearls seen in the market-
(average 261 ppmw) and light rim place (figure 15).
(average 23 ppmw). The elevated Mn ganese oxides would be incorporated Microscopic inspection revealed a
in the core is consistent with the dark into the opal and impart a dark body- surface structure that lacked the char-
color, as black manganese oxides have color. Qualitative analysis by EDXRF acteristic platelet structure of nacre-
been observed on Wollo opal. If Mn also showed relatively high levels of ous pearls. Instead, the surfaces had a
was present during opal formation, it Mn when compared to a white opal “glittery” painted appearance and
is reasonable to conclude that man- from Wollo. unnaturally distributed orient, fea-

Figure 13. This 10.67 ct opal Figure 15. This strand contains 17 large (up to 29.46 mm), coated,
cabochon is reportedly from the baroque-shaped, bead-cultured freshwater pearls.
Wollo Province of Ethiopia. The
black bodycolor is caused by
natural manganese oxides incor-
porated during formation.


Edition of the Gardner-Sward Hand-
book, American Society for Testing
and Materials, Philadelphia, 1995, pp.
229–230). Natural saltwater pearls
with a bismuth-bearing coating were
reported in a Fall 2005 Gem News
International entry (pp. 272–273).
This strand proved unusual for
two reasons: Bead-cultured freshwa-
ter pearls are not common in the mar-
ket, and the pearl coatings we have
examined have tended to consist of
Figure 16. Microscopic examination of the cultured pearls reveals air bubbles clear silicone polymers applied to
trapped in folds within the surface (left, magnified 20×). A close-up of a drill improve luster or protect the underly-
hole shows puckering and peeling of the outer surface, as well as the underly- ing nacre (see Lab Notes: Spring 2000,
ing nacreous surface of the cultured pearl (right, magnified 20×). p. 65; Spring 2002, pp. 83–84). The
coating on the present strand created
the illusion of thick, luminous nacre,
and was probably applied to reinforce
the precariously thin nacre layer on
tures commonly found in imitation MS chemical analysis of an exposed
these cultured pearls.
pearls. Also, air bubbles were trapped nacreous layer, which showed high
Akira Hyatt
in the folds and crevices throughout concentrations of Mn and Sr. Both
the strand (e.g., figure 16, left). EDXRF and LA-ICP-MS detected sig-
Puckering at the drill holes, as well as nificant amounts of bismuth in the Tenebrescent ZIRCON
noticeable chipping and peeling, fur- coating. Bismuth is often used in arti- Only a few gem varieties are known
ther indicated a pearl imitation. In ficial coatings to create a pearlescent to exhibit reversible photochromism,
addition, some of the drill holes appearance (J. V. Koleske, Ed., Paint also known as tenebrescence. This
revealed an underlying nacreous layer and Coating Testing Manual: 14th phenomenon is most notable in hack-
(figure 16, right). These observations manite a variety of sodalite. Recently
suggested the strand consisted of we learned about another gem show-
nacreous pearls with a thick coating, ing tenebrescence: zircon. Our source
which would explain their organic- Figure 17. X-radiography of the indicated that five such stones had
appearing shape despite their artificial strand reveals a bead-cultured been discovered approximately 25
appearance. structure and very thin nacre lay- years ago in central Australia. When
Most of the samples showed obvi- ers around the beads. left in the dark for a few hours, the
ous uniformly round bulges like those specimens reportedly turned orange,
seen in baroque bead-cultured pearls. which would fade to near-colorless
Examination of these areas with fiber- within a few minutes of exposure to
optic lighting showed distinct band- light (figure 18).
ing that is typical of shell beads. X- Two of the Australian samples
radiography revealed that the entire were provided for examination in
strand was indeed bead cultured (fig- September, 2011: a 3.07 ct piece of
ure 17). No distinct boundary rough and a 1.80 ct faceted round bril-
between the very thin nacre and the liant. Both samples had fairly typical
coating could be observed in the X-ray properties for zircon: RI—over the
images. limits of the refractometer (>1.81);
The long-wave UV reaction was SG—4.72; and fluorescence—very
relatively inert across the body of the weak orange to long-wave UV radia-
cultured pearls but strong bluish tion, and moderate orangy yellow to
white along the narrower areas where short-wave UV. Microscopic exami-
the coating was most concentrated. nation revealed reflective disk-like
When exposed to X-rays, the strand inclusions. LA-ICP-MS chemical
showed moderate yellow lumines- analysis did not show anything out of
cence (which was visible through the the ordinary for zircon.
coating), typical of a freshwater ori- When first taken out of the dark
gin. This was confirmed by LA-ICP- environment, the stones were orange.


temperatures if the stones are exposed
to light. . . . In some cases the colour
becomes pale. . . . Such altered stones,
if kept in darkness, will recover their
original colour. . . .” Interestingly, sev-
eral other references gave fairly simi-
lar descriptions of this phenomenon,
but it was never referred to as tene-
brescence—even though it fits the
definition. Likewise, a Fall 1986 Lab
Note (pp. 188–189) documented
reversible color-change in a blue zir-
con, which changed to an undesirable
grayish brown with exposure to long-
wave UV radiation.
Figure 18. This 1.80 ct tenebrescent zircon is shown just after it was Several online retailers are selling
taken out of the dark (left), and after exposure to a fiber-optic light for this material as tenebrescent zircon.
several minutes (right). The stones are reported to be from
Tanzania, Australia, Cambodia, or
Nigeria; one site says it was found 10
Placed under a standard 13.3 watt color of these zircons with short-wave
years ago. It remains unclear whether
incandescent desk lamp, they faded UV were unsuccessful. We also tried
any of this information is reliable,
within a couple of minutes to a light, low heat without success, but this
but one thing is certain: tenebrescent
slightly pinkish brown. The color did may be because the samples were not
zircon exists, and it is a wonder that
not appear to change any further under heated enough; they were not heated
this interesting material has been so
indirect office lighting. Additional further out of concern for potential
little reported in the contemporary
exposure to a 150-watt fiber-optic light damage.
for approximately one minute caused A search of the literature by GIA’s
Shane F. McClure
them to fade to pale pinkish brown. Richard T. Liddicoat Library and
Longer exposure to the fiber-optic light Information Center found no men-
did not cause any more fading. After tion of “tenebrescent” zircon,
being stored in the dark overnight, the although several books described zir-
Sood Oil (Judy) Chia—1; Paul Johnson—2
stones returned to their original color. cons with such behavior. The oldest
and 5; Jian Xin (Jae) Liao—3 and 15;
We repeated this process several times reference found was in Max Bauer’s Sally Eaton-Magaña—6; David Nelson—7;
with the same results. Precious Stones (1904), which noted: C. D. Mengason—8 and 18; Robison
Some tenebrescent stones change “Zircon—The colour and luster of McMurtry—11, 13, and 14; Nathan
back to their original color by expo- some hyacinths [probably referring to Renfro—12; JaeWon Chang—16.
sure to short-wave UV radiation or brownish or pinkish orange material]
heat. Attempts to restore the orange is liable to change even at ordinary

For online access to all issues of GEMS & GEMOLOGY from 1981 to the present, visit:


Brendan M. Laurs (

Contributing Editors
Emmanuel Fritsch, CNRS, Team 6502, Institut des Matériaux Jean Rouxel (IMN), University of Nantes, France (
Michael S. Krzemnicki, Swiss Gemmological Institute SSEF, Basel, Switzerland (
Franck Notari, GGTL GemLab–GemTechLab, Geneva, Switzerland (
Kenneth Scarratt, GIA, Bangkok, Thailand (


Chondrodite from Mahenge, Tanzania. At the 2011
Tucson gem shows, Dudley Blauwet (Dudley Blauwet
Gems, Louisville, Colorado) showed GIA some samples of
chondrodite that were found near Mahenge, Tanzania, in
early 2010. Chondrodite, (Mg,Fe2+)5(SiO4)2(F,OH)2, is a fair-
ly rare monoclinic silicate mineral that typically occurs
as small grains ranging from yellow to red and brown;
well-formed crystals and gem-quality material are rare.
Faceted chondrodite was previously reported from
Tanzania, though from a different locality, Sumbawanga
(see Winter 2007 Gem News International [GNI], pp.
Mr. Blauwet was aware of ~100 g of rough material
being produced at Mahenge, which consisted of small Figure 1. These chondrodites (left, 0.34 ct) are from
waterworn pebbles and larger broken fragments. He Mahenge, Tanzania. Photo by Robert Weldon.
bought a few small samples in 2010, and the cut stones
from this parcel yielded 121 pieces totaling 19.06 carats.
Later, during the 2011 Tucson shows, he purchased about
50 g of clean rough. This parcel produced 99 cut stones troscope. Inclusions consisted of numerous transparent
totaling 55.57 carats. It included five gems weighing >1.5 reflective particles and fine oriented short needles, some
ct; the largest one was ~2.0 ct. appearing rusty orange. Compared to the Sumbawanga
Gemological examination by GIA in Tucson yielded chondrodite, the Mahenge material had a slightly higher
the following properties from a 0.34 ct faceted sample (fig- RI, but the same birefringence. The needle-like inclusions
ure 1, left): RI—nα= 1.594, nβ = 1.602, nγ = 1.621; birefrin- were not seen in the Sumbawanga material.
gence—0.027; optic character—biaxial positive; pleochro- Mr. Blauwet reported that, to his knowledge, no addi-
ism—weak yellow to near colorless; UV fluorescence— tional gem-quality chondrodite has been produced from
inert to long-wave, and slightly chalky yellow to short- Mahenge since 2010.
wave UV; and no lines visible with the desk-model spec- Thomas W. Overton (
Carlsbad, California

Blue dolomite from Colombia. Dolomite, a calcium-mag-

Editor’s note: Interested contributors should send informa-
nesium carbonate [CaMg(CO3)2], is the second most
tion and illustrations to Brendan Laurs at or
GIA, The Robert Mouawad Campus, 5345 Armada Drive, important carbonate mineral after calcite. Although
Carlsbad, CA 92008. Original photos will be returned after dolomite is sometimes used as an ornamental stone, it is
consideration or publication. rarely seen as a gem due to its low hardness (Mohs 3½–4)
GEMS & GEMOLOGY, Vol. 47, No. 4, pp. 316–335, and lack of transparency. However, in late 2010, GIA was informed by Farooq Hashmi (Intimate Gems, Glen Cove,
© 2011 Gemological Institute of America New York) about a new find of gem-quality blue dolomite
from the famous Muzo emerald mine in Colombia. The


phase inclusions with jagged edges (figure 3, left), along
with other fluid and transparent crystalline inclusions.
Healed feathers containing tiny crystals were found in
most samples. A wavy strain pattern was observed with
cross-polarized light. Dolomite’s high birefringence (0.180)
was evident from the strong doubling. An albite inclusion,
identified by Raman and LA-ICP-MS analysis, broke the
surface of one rough sample (figure 3, right). Pyrite, com-
monly associated with dolomite at Muzo, was also seen
on the surface of a few rough samples, all of which were
rhombohedral twinned aggregates with reentrant corners.
The surfaces also showed numerous etch features. One
crystal contained three large, flat fractures parallel to
rhombic faces, consistent with the perfect rhombohedral
cleavage of dolomite.
Figure 2. This 1.13 ct oval brilliant is a rare blue A UV-Vis spectrum of the cut stone revealed Fe3+
dolomite from the Muzo mine, Colombia. Photo by absorption bands at 450, 460, and 470 nm, along with a large
Brad Payne. broad band centered at ~580 nm. Qualitative EDXRF spec-
troscopy detected Ca and Mg, as well as traces of Fe and
Mn. LA-ICP-MS analysis revealed many additional trace
material was reportedly produced in mid-2010 from veins elements, including Sc, Ti, V, Cr, Ni, Zn, Sr, Y, and rare-
in black shale. Less than 2 kg of gem-quality rough were earth elements. A pure dolomite crystal is expected to be
found, as crystals up to 2.5 cm across (mostly ~1 cm). Mr. colorless, and the blue color shown by this Colombian
Hashmi arranged for some of the rough material to be material may be due to natural radiation. Blue color in car-
donated to GIA by Greg Turner (Sacred Earth Minerals, bonate minerals (i.e., calcite) that contain twinning and dis-
Asheville, North Carolina), and Brad Payne (The Gem locations is thought to be related to natural radiation (T.
Trader, Surprise, Arizona) also loaned a 1.13 ct oval bril- Calderon et al., “Relationship between blue color and radia-
liant (figure 2) for examination. tion damage in calcite,” Radiation Effects, Vol. 76, 1983, pp.
All of the samples were transparent light blue without 187–191,
any color zoning. Their gemological properties were con- Transparent (colorless) dolomite has been documented
sistent with dolomite, which was confirmed by Raman from Spain (see M. O’Donoghue, Ed., Gems, 6th ed.,
analysis. Microscopic observation revealed two- and three- Butterworth-Heinemann, Oxford, UK, 2006, p. 406), but

Figure 3. The dolomite contained tiny jagged three-phase inclusions associated with other fluid inclusions
(left, magnified 110×). An albite inclusion broke the surface of one rough dolomite sample (right, magnified
80×). Photomicrographs by K. S. Moe.


Figure 4. This unusual
pendant, measuring ~3 ×
8 cm, contains a slab of
Namibian fluorite that is
surrounded by a piece of
brushed silver. It is
shown in reflected light
(left) and transmitted
light (right). Courtesy of
Goldideas, Windhoek;
photos by Jo-Hannes

we believe this is the first report of blue dolomite being Windhoek, Namibia) about a new find of fluorite located
used as a gemstone. several kilometers from Klein Spitzkoppe toward the
Kyaw Soe Moe ( and Wai L. Win Khan River. The material shows distinctive color zoning
GIA, New York and inclusions, and Mr. Brunner reported that only a
small amount has been mined so far. He indicated that a
Fluorite from Namibia. The Klein Spitzkoppe area in few dozen polished slabs have been produced in sizes up
Namibia is known as a source of attractive crystals of to 10+ cm in longest dimension, and some of them have
topaz, aquamarine, and other minerals from miarolitic been set in silver pendants (e.g., figure 4). Mr. Brunner
cavities (see, e.g., B. Cairncross et al., “Topaz, aquama- donated to GIA one polished slab (89.47 ct, or 37.05 ×
rine, and other beryls from Klein Spitzkoppe, Namibia,” 39.75 × 5.60 mm) and seven rough pieces of this fluorite
Summer 1998 G&G, pp. 114–125). In mid-2011, GIA (9.83 to 45.97 g) for examination.
was informed by Jo-Hannes Brunner (Pangolin Trading, Gemological examination of the slab revealed the fol-
lowing properties: RI—1.434; hydrostatic SG—3.12;
Chelsea filter reaction—none; fluorescence—inert to long-
Figure 5. Color zoning and several prismatic inclu- wave UV radiation, and weak yellow to short-wave UV in
sions containing dickite are visible in this Namibian the yellow to orangy yellow portions; and no clear absorp-
fluorite slab (gift of Jo-Hannes Brunner; GIA tion lines visible with the desk-model spectroscope. These
Collection no. 38388). Photomicrograph by C. Ito; properties are consistent with fluorite, and the identity of
magnified 16×. all the samples was confirmed by Raman spectroscopy.
The fluorite was distinctly color zoned, with cubic vio-
let zones inside yellow to orangy yellow areas. Whitish,
prismatic inclusions with generally square cross-sections
(figure 5) appeared opaque in transmitted light and were
predominantly hosted by the violet areas of the fluorite.
These inclusions were identified by Raman spectroscopy
as dickite, a clay mineral with the formula Al2Si2O5(OH)4.
Many of them consisted of partially hollow tubes, appar-
ently created when some of the soft dickite weathered
away or was removed during the polishing of the slab. A
few yellow areas within the dickite inclusions were identi-
fied as sulfur, and Raman analysis also detected quartz
inclusions in the fluorite. In addition, microscopic exami-
nation revealed reflective, iridescent fluid inclusions with
geometric patterns.
LA-ICP-MS analysis of both the yellow and violet por-
tions of the fluorite showed trace amounts of Ti, Sr, La,
and Ce. The violet area tended to show higher concentra-


Figure 6. Shown here
are rough and cut
opals from Laverton,
Western Australia.
The faceted stones
weigh 0.50–4.39 ct.
Photo by V. Pardieu.

tions of these elements, as well as trace amounts of addi- Central Road at coordinates 28°22′10′′ S, 122°35′59′′ E.
tional rare-earth elements. According to local geologic maps, this region consists
UV-Vis-NIR spectroscopy of a yellow portion of the mainly of weathered biotite monzogranite or kaolinized
slab showed a broad band at 434 nm, while a violet section granites. The opal seam appears to be associated with a
displayed a 306 nm peak with broad bands at approximate- fault trending northeast/southwest. In September 2008,
ly 410 and 570 nm. The band in the yellow region is con- Holdfast Exploration was granted a five-year exploration
sistent with the “yellow center” attributed to an O3– ion license, and prospecting has been carried out since
replacing two adjacent F– ions (H. Bill and G. Calas, “Color November 2008. Pits were excavated up to about 1.5 m
centers, associated rare-earth ions and the origin of col- deep in six different areas using mainly hand tools, a jack-
oration in natural fluorites,” Physics and Chemistry of hammer, and an excavator.
Minerals, Vol. 3, 1978, pp. 117–131). The violet portion Samples of opal in host rock, loose pieces of rough, and
showed an absorption spectrum similar to those of purple faceted stones were donated to GIA in December 2010.
fluorites in that article, with a 570–580 nm band (possibly Eight samples weighing 0.49–4.39 ct studied for this report
attributed to colloidal calcium) and other possible Y- or represented the color range of the Laverton material: two
Ce-associated F-center features. rough specimens (yellow and brownish orange) and six
Absorption spectra of the fluorite and Raman spectra of faceted stones (two colorless, two yellow, and two brown-
the inclusions are available in the G&G Data Depository ish orange). Some of the matrix specimens contained col-
at The delicate color patterns and interest- orless, yellow, and brownish orange opal within the same
ing inclusion scene displayed by this fluorite makes it an piece. Most of the opal in matrix showed some unhealed
attractive option for jewelry use when cut as slabs. surface-reaching fissures that were present before the
Claire Ito ( stones were brought to GIA. Observations over an 11-
GIA, New York month period showed some evidence of crazing in one of
them, the colorless round brilliant shown in figure 6.
Common opal from Laverton, Western Australia. In early Gemological properties of the opal are summarized in
2011, Peter Piromanski from Holdfast Exploration Pty. table 1. Most of the samples showed some turbidity and
Ltd., Wannero, Western Australia, showed this contributor flow patterns. One of the two colorless pieces contained
some attractive potch opals (e.g., figure 6), to be marketed clusters of minute crystals (figure 7, top left). In most of the
as Piroman Opal, that were reportedly from a new deposit yellow material, small spheres were seen individually (fig-
31 km north of the town of Laverton, near the Great ure 7, top right) or in groups (figure 7, bottom left). In the


Revue de Gemmologie, No. 138–139, 1999, pp. 34–40).
TABLE 1. Properties of opal from Laverton, Raman spectroscopy of the same five samples showed a
Western Australia. rather broad asymmetric band centered at about 350 cm−1
Color Colorless to saturated yellow and with smaller bands at 1220, 1075, 965, and 780 cm−1. Such
brownish orange. Some stones showed spectra are typical of opal-CT (M. Ostrooumov et al.,
orange and colorless color zoning. “Spectres Raman des opales: Aspect diagnostique et aide à
Diaphaneity Transparent to translucent la classification,” European Journal of Mineralogy, Vol. 11,
RI 1.40–1.44 1999, pp. 899–908), which is usually found in a volcanic
SG 1.99–2.01 setting.
Mohs hardness 5–6 The discovery of fire opal in Western Australia, a part
UV fluorescence Yellow and brownish orange opal: inert of the country not usually associated with opal, is an inter-
Colorless opal: strong white (with 2–3
esting development. Prospecting is ongoing, and updates
seconds of phosphorescence) to short-
wave UV and additional information on this opal will be posted at
Spectroscope spectrum Colorless to yellow opal: no features These attractive gems could
Brownish orange opal: strong absorption make a welcome addition to Australian opal production.
in the blue region Vincent Pardieu (
GIA, Bangkok

brownish orange opal, the most common inclusions were A bicolor, bi-pattern hydrophane opal. The Laboratoire
tiny angular, often reddish crystals associated with ran- Français de Gemmologie in Paris recently examined an
domly oriented reflective (sometimes iridescent) discoid unusual 17.15 ct opal cabochon, measuring approximately
tension fissures, reminiscent of the “lily pads” seen in peri- 20.95 × 15.96 × 11.69 mm. Play-of-color in the full rain-
dot (figure 7, bottom right). We have been unable to identi- bow of hues was visible throughout the stone (figure 8,
fy any of these inclusions with Raman microspectroscopy. left), but the bodycolor was not even. One end of the cabo-
EDXRF analysis of five opals (colorless, yellow, and chon was medium brown, while the rest of the gem had a
brownish orange) showed traces of Cu and Zr in all sam- translucent whitish color (figure 8, right). The color
ples, while Ca, Fe, and Sr were detected in all but the color- boundary formed a well-defined line along the base of the
less pieces. The three darker brownish orange stones con- stone. Such a zoning pattern has been occasionally seen in
tained the highest Fe contents, as expected from the litera- common opal but is unusual for play-of-color opal.
ture (E. Fritsch et al., “Découvertes récentes sur l’opale,” To better characterize the opal, we weighed it after 12

Figure 7. Inclusions seen

in the Laverton opals
included a cluster of
minute unknown crys-
tals (top left) and tiny
spherical inclusions (top
right), which sometimes
occurred in clusters (bot-
tom left). Randomly ori-
ented crystals associated
with lily pad–like ten-
sion fissures were also
seen (bottom right).
Photomicrographs by V.
Pardieu, magnified ~40×.


Figure 8. This 17.15 ct bicolor, bi-pattern opal is likely from Ethiopia’s Wollo Province. The opal’s color zon-
ing is best viewed from the side. One end has a translucent whitish color, and the other is medium brown.
Photos by A. Droux.

hours of drying in air in the laboratory, and again after stronger fluorescence than the rest of the opal.
immersion in water. Water fills the pores of opals that Another peculiarity was that the two color zones
show hydrophane character. After immersion the stone showed different patterns in their play-of-color. The brown
weighed 17.99 ct, which clearly indicated a hydrophane zone had fairly large patches and a striated appearance,
character, but its appearance remained unaltered. sometimes referred to as a “straw” or “chaff” pattern (fig-
To avoid introducing foreign substances into the ure 9). This feature is due to polysynthetic twinning of the
porous opal, we did not test for RI or SG. The gem emitted network of silica spheres found in many opals, particularly
a weak whitish fluorescence to long-wave UV radiation, those from Ethiopia. This pattern was absent from the
with a slightly less intense reaction to short-wave UV. The light-colored portion, which had smaller patches, less-visi-
emission was zoned, with the brown portion nearly inert. ble borders, and somewhat “rolling” color flashes.
The cap of the white zone at the top of the cabochon had a The base of the cabochon showed a cellular pattern of

Figure 9. The “straw” or “chaff”

pattern in the brown zone is due
to polysynthetic twinning of the
network of silica spheres consti-
tuting the opal. Photomicrograph
by A. Droux; magnified 55×.


Figure 10. This 4.56 ct quartz cabochon displayed Figure 12. This surface-reaching cinnabar inclusion
vivid red coloration and subtle chatoyancy. Photo by shows a six-fold euhedral shape in cross-section. The
B. Rondeau. upper left portion of the inclusion has broken away,
and the original trace is outlined. Image by B. Rondeau.

well-formed “digits” that is typical of some opals from

Wollo, Ethiopia, along with the abundance of twinning in
the brown portion of the stone. Whitish play-of-color opal Chatoyant quartz with cinnabar inclusions. While on a
is common from Wollo but not from Ethiopia’s Shewa buying trip to Jaipur, India, one of these contributors (TP)
area. We concluded that this gem is likely from the Wollo purchased a 4.56 ct cabochon of what appeared to be
deposits. quartz with iron oxide inclusions, such as those seen in
Alexandre Droux “strawberry” quartz from Kazakhstan. The cabochon mea-
Laboratoire Français de Gemmologie, Paris sured 10.4 × 8.7 × 6.8 mm and had a vitreous luster (figure
10). Its spot RI of ~1.54, SG of 2.64, and inert reaction to
Emmanuel Fritsch
long- and short-wave UV were all consistent with quartz.
The gem was vivid red under reflected light except for a
narrow linear near-colorless band. Most interestingly, it
Figure 11. The chatoyancy or star effect in the quartz displayed chatoyancy in fiber-optic light, appearing as a
is due to numerous oriented short needles of cinnabar. four-rayed star in some positions. Microscopic examina-
Photomicrograph by B. Rondeau; magnified 50×. tion revealed that the phenomenon was due to numerous
red inclusions. These were short (<1 mm) acicular crystals
oriented mainly along four directions (figure 11). When
viewed with a scanning electron microscope (SEM), they
showed a euhedral shape, with six-fold symmetry in cross-
section, and were about 3 µm wide in narrowest dimen-
sion (figure 12). Since their morphology did not resemble
that of iron oxide, we investigated them further.
Chemical analysis of surface-reaching inclusions using
a JEOL 5800 SEM equipped with an energy-dispersive spec-
trometer detected only mercury and sulfur, a composition
consistent with cinnabar. This identification is also consis-
tent with their hexagonal symmetry. Cinnabar inclusions
in quartz have been documented previously, especially in
material from China (J. Hyršl and G. Niedermayr, Magic
World: Inclusions in Quartz, Bode Verlag, Haltern, Ger-
many, 2003, p. 53; E. J. Gübelin and J. I. Koivula, Photo-
atlas of Inclusions in Gemstones, Vol. 3, Opinio Verlag,
Basel, Switzerland, 2005, pp. 592, 627). To the best of our


“tourmalinated” or “rutilated” quartz, respectively.
Recently, the Gem Testing Laboratory in Jaipur examined a
quartz specimen that contained eye-visible emerald crystals
(figure 13). Although intergrowths of emerald and quartz
and a notable emerald-in-quartz specimen have been report-
ed previously (e.g., Lab Notes: Summer 2000, pp. 164–165;
Fall 2008, p. 258), this was quite different.
The slightly smoky 45.85 ct marquise-shaped cabochon
measured 37.60 × 18.26 × 11.57 mm. The prominent green
inclusions displayed an acicular habit (figure 14, left). Their
green color and hexagonal profile (figure 14, right) strongly
suggested emerald, but their acicular habit raised some
doubts, as emeralds typically show a more columnar form.
Most of the crystals also displayed basal parting planes,
reminiscent of the actinolite blades found in emeralds from
the Ural Mountains of Russia. Some also displayed color
Figure 13. This 45.85 ct quartz cabochon was unusu- zones following the prism faces, while others contained
al for its eye-visible inclusions of elongated emerald rain-like inclusions.
crystals. Photo by G. Choudhary. To conclusively identify the inclusions, we examined
the sample under a desk-model spectroscope. It revealed a
spectrum consistent with emerald, featuring a doublet in
the red region and an absorption band in the yellow-green
knowledge, however, cinnabar has not been reported as ori- region. Further confirmation was obtained by FTIR,
ented inclusions responsible for chatoyancy or a star effect, which displayed a typical emerald spectrum. IR spec-
which makes this an unusual and attractive specimen. troscopy also confirmed the host material as quartz,
Thierry Pradat ( which was supported by a spot RI of 1.54 and a hydrostat-
G-Plus, Lyon, France ic SG of 2.65.
Benjamin Rondeau Textural relationships indicated that the emerald crys-
Laboratoire de Planétologie et Géodynamique, tals formed before the host quartz (i.e., they are protoge-
CNRS, Team 6112, University of Nantes, France netic). Emerald is known to occur within quartz, but this
Emmanuel Fritsch sample was quite unusual for the crystals’ acicular habit
and their occurrence as inclusions, not merely in associa-
Quartz with acicular emerald inclusions. Quartz with ran- tion with the quartz or as an intergrowth.
domly distributed tourmaline or rutile needles is widely Gagan Choudhary (
available in the market. These gems are often described as Gem Testing Laboratory, Jaipur, India

Figure 14. The inclusions displayed an acicular habit (left, magnified 32×), which is usually not associated with
emeralds. However, their green color and hexagonal profile (right, magnified 48×) helped identify them as emer-
ald. (The green material in some quartz fractures is polishing powder.) Photomicrographs by G. Choudhary.


Figure 15. This life-size
quartz skull carving con-
tains inclusions of the rare
mineral izoklakeite.
Courtesy of Harold Van Pelt;
photo by Erica Van Pelt.

Quartz carving with inclusions of izoklakeite. These con- percussion mortar. This freed numerous fragments of the
tributors recently examined a life-size quartz skull (figure silver-gray included crystals. Powder X-ray diffraction
15) carved by noted gem photographer and lapidary Harold analysis at the Natural History Museum of Los Angeles
Van Pelt (Los Angeles). The hollowed skull contained two County pointed to either izoklakeite or giessenite, and an
articulated pieces and a jaw hinge that opened and closed. electron back-scattered diffraction pattern done at Caltech,
Starting with a 250 lb. colorless quartz crystal, Mr. Van
Pelt produced this 6.5 lb. (2.9 kg) sculpture.
One of the carving’s interesting features was its abun- Figure 16. Izoklakeite and possibly other sulfides
dance of conspicuous submetallic silver-gray inclusions form these inclusions in the quartz used in the carv-
(e.g., figure 16). These were initially believed to be ing from figure 15. Photomicrograph by G. R.
jamesonite, a lead-iron-antimony sulfide mineral often Rossman; image width 3.2 mm.
found in fibrous form. Initial analyses with a scanning elec-
tron microscope (SEM) quickly determined otherwise. No
iron was detected, which meant the fibers consisted of a
different material.
A more detailed SEM and electron microprobe investi-
gation at Caltech revealed a lead-antimony-bismuth sulfide
with a minor amount of copper. Its formula was initially
determined as (Pb2.65Cu0.25)(Sb1.14,Bi0.95)S6. Areas of some of
the fibers also consisted of several alteration products: gale-
na (PbS), bismuthinite (B2S3), some CuPb(Sb,Bi)S3 minerals,
and small amounts of native bismuth.
To identify the original lead-antimony-bismuth sul-
fide, we obtained a ~1 cm3 piece of the quartz from which
the skull was originally carved. It was cooled in liquid
nitrogen to make it brittle and immediately shattered in a


Figure 17. A Pakistani miner breaks marble to collect rubies at Datumbaresho, in the Hunza Valley.
Photo by V. Pardieu.

together with the chemical analyses, confirmed that the (Syed Trading Co., Peshawar, Pakistan), we traveled first to
phase was izoklakeite. northern Pakistan to visit ruby deposits near Hunza and
Izoklakeite is the antimony member of a solid-solution Bisil. Author VP, with assistance from Zulfiqar Ali Abbas
series with giessenite, the bismuth-dominant member. It (Kashmir Gems Ltd., Abbottabad, Pakistan), then proceed-
was first described from a sulfide body near Izok Lake, ed to the ruby and sapphire deposits in the Kaghan Valley
Canada (D. C. Harris et al., “Izoklakeite, a new mineral near Batakundi. This report provides an update on the
species from Izok Lake, Northwest Territories,” Canadian mining and production at some of the Pakistani deposits
Mineralogist, Vol. 24, 1986, pp. 1–5). The mineral was described in the Fall 2007 (pp. 263–264) and Winter 2010
later found in Sweden and as inclusions in quartz crystals (pp. 319–320) GNI entries.
from Switzerland. Its ideal chemical formula is About 100 miners and local dealers were involved with
Cu4Pb54Sb38S114. ruby mining and trading around the Hunza Valley, at sev-
The quartz skull, named “Izok” in honor of its unusual eral deposits that initially started producing in the 1960s
inclusions, is part of an exhibit of Harold Van Pelt’s carv- and 1980s. The rubies are found in marbles, sometimes
ings on display at the Houston Museum of Natural associated with mica and blue and pink spinel. The
Science through October 2012. deposits are located north of the valley, from Datum-
George R. Rossman ( and Chi Ma baresho (figure 17) in the northwest to the Aliabad and
California Institute of Technology Karimabad areas (Bajoring, Gharei Chhar, Phudan Daar,
Pasadena, California Gafinas), Altit, Ahmedabad, and Dong-e-Das (also known
Anthony R. Kampf as Ganesh) in the east. Most of the workings are located
Natural History Museum of Los Angeles County 500–1,000 m above the villages at an elevation around
3,000 m (nearly 10,000 feet), and they are difficult to
Update on ruby and sapphire mining in Pakistan. In access because of the steep terrain. The color of the ruby
August and September 2011, these contributors visited generally ranges from deep red at Datumbaresho to pink-
Pakistan to collect reference samples for GIA’s laboratory. ish red at Aliabad/Karimabad to pinkish and bluish red
With the support of gem merchant Syed Iftikhar Hussein around Dong-e-Das. The size of the stones appears to


tools; production is extremely limited.
Also in the Batakundi area, the Besar (also known as
Basil or Besel) sapphire mines were being worked by about
50 miners using jackhammers and explosives at two sites,
located at 35°02′41′′ N, 73°52′56′′ E, and 35°02′58′′ N,
73°53′19′′ E (e.g., figure 19). The elevation of these sites is
4,020 m and 3,780 m, respectively. Several kilograms of
pink to purple sapphires were being extracted from the
graphite-rich veins on a daily basis; less than 5% was of
gem quality.
The overall ruby production from Pakistan is very
low compared to other primary-type deposits in Asia and
Africa. Most of the ruby production in Central Asia
comes from the Murgab area in eastern Tajikistan and
from Jegdalek, Afghanistan. Pakistani rubies are typically
more included and smaller. Although Pakistan’s ruby
mines are rarely affected by the political instability and
security issues present elsewhere in the country, the
deposits are located high in mountainous areas that are
difficult to access.
Vincent Pardieu
Stephane Jacquat
Piat Co., Bangkok

Figure 19. At the Besar deposits, in the Batakundi

area of Pakistan’s Kaghan Valley, several miners use
Figure 18. These rubies are from the Bisil area, in the jackhammers and explosives to break the hard rock
Basha Valley of Pakistan. The largest piece, on the far in search of sapphires. Photo by V. Pardieu.
right, is ~1 cm long. Photo by V. Pardieu.

increase in that same direction, from <0.4 g to 20+ g. Most

Hunza ruby is cabochon quality at best, though mineral
specimens are also important. Much of the region’s pro-
duction is apparently exported to China.
The Bisil ruby deposit is located near 35°52′53′′ N,
75°23′52′′ E, in the Basha Valley north of Skardu. It was
discovered in 2003 by a local prospector. Rubies from Bisil
(figure 18) are also found in marble, within veins or pods,
typically in association with diopside and mica. We saw
about 10 mining pits, up to 20 m deep. The deposits
appear to extend from just above Bisil village (elevation
~2,800 m) to the top of the mountains at an elevation
above 3,500 m, and may extend more than 10 km in an
east-west direction. About 15 miners were working the
area using simple tools, jackhammers, and dynamite.
Small stones ranging from pink to deep red are mined
year-round, but production is very limited.
The Batakundi ruby deposit is located in the Kaghan
Valley at 34°52′41′′ N, 73°48′05′′ E, at an elevation of 4,190
m. Marble-hosted rubies, most of them small and deep
red, have been recovered there since 2000. Rough stones
larger than 0.6 g are very rare, with most of the production
smaller than 0.2 g. A group of seven miners led by Haider
Ali have been working in the current area since 2006. As
at Bisil, the miners use a jackhammer, dynamite, and hand


Figure 20. Sphene from the Pakistan/Afghanistan border area may contain a significant amount of V, which is
presumably responsible for its color-change behavior. The oval gem on the left (2.76 ct) shows a noticeable
color change from mainly yellowish green in daylight-equivalent lighting (left) to brownish orange in incandes-
cent light (right). Photos by Robert Weldon.

Vanadium-bearing color-change sphene from Pakistan/ more than 2400 ppmw V (and no Cr), which is presum-
Afghanistan. Recently seen in the Carlsbad laboratory ably the cause of the color change. The visible spectrum
was a parcel of rough and cut sphene (e.g., figure 20) showed a broad absorption feature centered at approxi-
loaned for examination by Eric Braunwart (Columbia mately 603 nm (figure 22). Transmission windows on
Gem House, Vancouver, Washington). According to him, either side of this broad band are consistent with the spec-
the sphene originated near the Pakistan/Afghanistan bor- trum expected for a color-change gem.
der. The most interesting feature of this material was that The coloration of this sphene is much different from
many of the stones showed a slight to moderate color the typically “golden” orange material previously
change. In daylight-equivalent lighting, the sphene was described from Pakistan’s North West Frontier Province
dominantly vivid green to yellowish green, changing to (Spring 2006 GNI, pp. 68–69), and also from the yellow
brownish orange or brown under incandescent light sphene known from Badakhshan, Afghanistan (Summer
(again, see figure 20). However, strong pleochroism was 2006 GNI, pp. 180–182).
responsible for the multiple colors seen in the faceted Nathan Renfro (
stones. GIA, Carlsbad
The rough consisted of well-formed blade-like crys-
tals, so it is apparent that the material was mined from a
primary deposit. Rough production as of July 2011 was
estimated by Mr. Braunwart to be around 600 g, from Figure 21. The most notable inclusions in the sphene
which approximately 200 carats of finished material have were crystals of near-colorless apatite and dark elon-
been produced. The cut stones ranged from calibrated gated amphibole. Also present were numerous fluid
sizes as small as 3 mm in diameter up to larger stones fingerprints. Photomicrograph by N. Renfro; magni-
weighing several carats. The largest faceted stone from fied 35×.
the production so far was a 5.45 ct oval brilliant.
Gemological properties of the material were consis-
tent with sphene. The RI was over-the-limit of the refrac-
tometer, and the average SG (measured hydrostatically)
was 3.54. A strong red reaction was observed with a
Chelsea filter. Microscopic observation revealed strong
doubling and numerous included crystals. These were
identified by Raman microspectroscopy as transparent
near-colorless apatite and dark elongated needle-like crys-
tals of an amphibole mineral (figure 21). Several stones
also contained fluid “fingerprints.” A few rough pieces
showed prominent zoning, and the color change was less
apparent across these zoned areas.
EDXRF spectroscopy of all samples showed major
amounts of Ca, Ti, and Si that are expected for sphene, as
well as traces of V, but no Cr. LA-ICP-MS measurements
of several spots on a faceted sample detected an average of






400 500 600 700


Figure 22. The visible spectrum of the color-change

sphene showed a broad absorption feature centered
at 603 nm that is presumably due to V, as this was
the only significant trace element detected by LA-
ICP-MS analysis.

Cobalt blue–colored spinel from Khuoi Ngan, Vietnam.

Blue spinel has been reported from several localities in the
An Phu–Luc Yen area of northern Vietnam, though highly
saturated material is quite rare (e.g., C. P. Smith et al., “A
closer look at Vietnamese spinel,” InColor, Spring 2008,
pp. 11–13; J. B. Senoble, “Beauty and rarity—A quest for
Vietnamese blue spinel,” InColor, Summer 2010, pp. Figure 23. At Khuoi Ngan, near An Phu in northern
18–23; Spring 2011 GNI, pp. 60–61). Testing has shown Vietnam, bright blue spinel is mined from shallow
that the blue color is due to various amounts of cobalt and pits by local villagers. Photo by Dudley Blauwet.
iron impurities (Smith et al., 2008).
Gem and mineral dealer Dudley Blauwet recently vis-
ited one of the Vietnamese blue spinel deposits and pro-
vided some information on the mining and production of name for Khuoi Ngan, as it refers to the same area of rice
this material. The site was located just east of the village fields). The largest piece of rough Mr. Blauwet purchased
of Khuoi Ngan and approximately 2.5 km southeast of An yielded a faceted oval weighing 0.80 ct (figure 25). During
Phu, where a few local villagers were mining secondary his latest visit to Vietnam, in November 2011, he reported
deposits in shallow pits along the edges of rice paddies (fig- very limited production of the intense blue spinel from
ure 23). Work at the site began in 2008, and a number of Khuoi Ngan, in small sizes that would cut stones expected
pits were excavated with hand tools to a depth of 2–3 m. to weigh <0.15 ct.
The gem-bearing material was taken to a nearby area for Mr. Blauwet mentioned that buyers continue to see
washing and hand picking (figure 24). glass and synthetics in rough and cut gem parcels in Viet-
The spinel rough was quite small but intensely col- nam. After receiving his first shipment of blue spinels from
ored. Mr. Blauwet obtained four small parcels ranging the cutting factory, a 0.26 ct sample was found to be glass.
from 0.4 to 5 g over the course of four days in Luc Yen. In Such imitations can be difficult to identify in the field
addition, he purchased a fifth lot of rough in An Phu on since the rough material commonly has been tumbled to
the day that he visited Khuoi Ngan. The individual pieces give the appearance of alluvial pebbles, which also disguis-
of rough were mostly <0.2 ct but suitable for cutting es any obvious bubbles or swirl marks. He reported that a
melee because of the high color saturation. All of the lots Chelsea filter is helpful in such situations: Co-bearing
were said to be from Khuoi Ngan, except one that was spinel appears pink to red, while glass shows no reaction.
reportedly from Bai Ruong (which may be an alternative Three faceted samples of Mr. Blauwet’s intense blue


Figure 24. The spinel-bearing soil from Khuoi Ngan is washed and carefully sorted by hand to extract the small
pieces of spinel. Photo by Dudley Blauwet.

Khuoi Ngan spinel (0.24–0.80 ct) were examined at GIA’s consistent with spinel, except for an inert reaction to UV
Carlsbad laboratory. SG values ranged from 3.50 to 3.57, radiation. Raman spectroscopy confirmed the spinel iden-
and the RI of the largest stone was 1.716. LA-ICP-MS anal- tification. Six arms were easily recognized in each tablet,
yses showed 60–290 ppmw Co. Iron was 7,400–10,500 though the trapiche structure was not well defined.
ppmw, and significant traces of Zn, Ga, Ni, Mn, V, and Li Rather than a core, the 3.64 ct sample contained a point
also were recorded. As expected, all three samples where the arms intersected. The 6.76 ct tablet had an
appeared red in the Chelsea filter.
Thomas W. Overton
Figure 25. These spinels from Khuoi Ngan weigh up
Andy H. Shen
to 0.80 ct. Photo by Robert Weldon.
GIA, Carlsbad

Trapiche spinel from Mogok, Myanmar. Trapiche growth

structure, most commonly observed in emerald, has also
been seen in ruby, sapphire, tourmaline, quartz, and
andalusite. In January 2011, Bill Larson (,
Fallbrook, California) loaned five samples to GIA that were
represented as trapiche spinel from Mogok, Myanmar.
They were polished into round tablets that ranged from 1.2
to 3.0 cm, and all showed a subtle radiating texture that
was more visible with transmitted light. Two pieces (3.64
and 6.76 ct) were characterized by this contributor for this
The samples were translucent and ranged from pink-
ish orange to red with “arms” that were dark gray (e.g.,
figure 26). Gemological properties of both samples were


Figure 26. This 6.76 ct
tablet of trapiche spinel,
reportedly from Mogok,
may represent a spinel
pseudomorph after ruby.
The trapiche structure is
not well defined, but the
arms are easily recognized,
particularly in transmitted
light (left). Photos by
Robert Weldon.

indistinct core that was cloudy and full of cracks. The The spinels had not been exposed to heat treatment, as
arms were created by concentrations of gray wispy clouds confirmed by their PL spectrum, which featured a sharp
and a denser network of cracks than in the surrounding band at 685 nm.
material. Similar samples of trapiche spinel from Myanmar were
In both tablets, tiny black graphite inclusions (identi- characterized by M. Okano and A. Abduriyim (“Trapiche
fied by Raman analysis) were visible throughout with a spinel,” Gemmology, Vol. 41, No. 485, Issue 2, 2010, pp.
gemological microscope. In the 3.64 ct tablet, minute 14–15 [in Japanese]), who suggested that the trapiche struc-
inclusions of molybdenite, magnesite, goethite, and ture formed during the growth of tabular spinel crystals. A
hematite were detected in the arms (e.g., figure 27; see the pseudomorphic origin of this trapiche spinel after ruby is
G&G Data Depository for more images). Hematite crys- also possible, and is supported by the remnant ruby inclu-
tals were found between the arms, and hematite, magne- sion detected in one of the samples.
site, molybdenite, and calcite were found along the outer Editor’s note: Consistent with its mission, GIA has a
rim of the samples. In the 6.76 ct tablet, dolomite crystals vital role in conducting research, characterizing gem-
were detected in the core and the arms; goethite crystals stones, and gaining knowledge that leads to the determina-
were identified between the arms and in the rim. Notably, tion of gemstone origins. The samples studied in this
Cr3+ photoluminescence bands at 692 and 694 nm, which report are not subject to the Tom Lantos Block Burmese
are characteristic of ruby, were detected in one tiny includ- JADE Act of 2008, and their import was in accordance
ed crystal along the rim of this sample (figure 28). with U.S. law.
Numerous cracks and cavities were observed through- Kyaw Soe Moe
out the tablets. DiamondView images revealed bright blue
and purple zones in the cracks, while the host spinel
showed a faint blue color. In the 3.69 ct tablet, orange-red SYNTHETICS AND SIMULANTS
stains were seen with the microscope inside the fractures “Cat’s-eye pearls”: Unusual non-nacreous calcitic pearl
between the arms. Qualitative EDXRF analysis revealed imitations. Imitations of non-nacreous pearls are abundant
Mg, Al, Ti, V, Cr, Fe, Zn, Ga, Ca, and Mn in both samples. in the market. They can be made of just about any materi-

Figure 27. Micro-inclusions in the trapiche spinel identified using Raman spectroscopy included a molybdenite
crystal along the rim (left), a rhombohedral magnesite crystal in an arm (center), and a dark-appearing calcite
crystal along the rim (right). Raman microscope photomicrographs by K. S. Moe.


Figure 28. Photolumi-
nescence spectroscopy
detected remnants of
ruby in the 6.76 ct
trapiche spinel. In the
Raman microscope
image (inset, by K. S.
Moe), the light reflect-

ed from region A is
slightly different from
region B. Interestingly,
region A showed ruby
PL bands at 692 and
689 694 nm in addition to
spinel PL bands at
a 685, 687, and 689 nm;
region B showed only
spinel PL bands, as
640 660 680 700 720 740 recorded at several
other spots on the
WAVELENGTH (nm) sample.

al but are usually sculpted from shell. Shells thick enough making them easy to identify.
to yield such pearl imitations are rather common, though This contributor recently received for testing six
generally they are restricted to aragonitic mollusks such as unusual black “pearls” (figure 29) from two different
Strombus sp., Tridacna sp., and Cassis sp. Calcitic shells, clients. Represented as non-nacreous natural pearls, the
including those of Pinna sp. and Atrina sp., tend to be thin pieces weighed between 2.52 and 10.26 ct and exhibited
and are therefore not suitable. Shell imitations show a lay- chatoyancy when illuminated with an intense light
ered structure when viewed with strong transmitted light, source. The chatoyancy was displayed all around the

Figure 29. Represented as

non-nacreous natural
pearls, these sculpted
pieces of shell (2.52–10.26
ct) exhibit distinct chatoy-
ancy. Photo by T.


Figure 30. The sculpt-
ed shell’s fibrous
structure is clearly
visible with magnifi-
cation (left). The thin
polished section on
the right displays a
layered appearance in
transmitted light that
is expected for shell.
Photomicrographs by
T. Hainschwang.

samples when they were rotated about an axis perpendic- Large synthetic quartz. Recently, the Gübelin Gem Lab
ular to the band, but it was not visible along this axis. received four large gems (figure 31) submitted for analysis
The phenomenon was caused by light reflections from by Erwin Walti (Oetwil am See, Switzerland). They were
their fibrous structure (figure 30, left). Viewed from the represented as amethyst (surprising for a dark green gem),
side, no fibrous structure was visible, but hexagonal out- aquamarine (greenish blue and light blue concave cuts),
lines of the fibers could be seen. These parallel layers of and citrine (yellow oval cut), and they weighed 162.00,
very fine hexagonal prisms are characteristic of calcitic 33.20, 36.30, and 38.40 ct, respectively. To improve work-
pearls and shell. As expected, reflectance infrared spec- flow, the laboratory occasionally puts stones through more
troscopy identified the samples as calcite. Distinct con- advanced analyses, such as FTIR and EDXRF spec-
centrations of Sr, detected by EDXRF chemical analysis, troscopy, prior to gemological testing. Such was the case
confirmed that they were of marine origin. with these samples.
The surface of the samples showed obvious polishing We noticed immediately that the analytical data did
lines, a common feature in polished and worked non- not match either natural quartz or aquamarine. All four
nacreous natural pearls. The telltale layered structure of FTIR spectra showed total absorption below approxi-
sculpted shell was not visible since the samples were mately 3600 cm−1 and only a weak, broad absorption
practically opaque to strong fiber-optic light. X-radio- band at 5196 cm−1 (e.g., figure 32). These IR features are
graphs did not reveal any structure, which is very unusu- consistent with those reported for some synthetic quartz
al for brown to black calcitic natural pearls; those from (e.g., P. Zecchini and M. Smaali, “Identification de l’orig-
Pinnidae mollusks typically show distinct concentric ine naturelle ou artificielle des quartz,” Revue de Gem-
structures. mologie a.f.g., Vol. 138–139, 1999, pp. 74–83). EDXRF
Because the orientation of the prismatic fibers and the spectroscopy of all four stones showed major amounts of
resulting chatoyancy clearly indicated a layered structure, Si with traces of Fe. The dark green sample also con-
and none of the properties were consistent with natural tained small but significant amounts of cobalt (0.005
pearls, these objects could only be sculpted shell. To more wt.% Co3O4), indicative of synthetic origin, while the
closely examine the layered structure of these “cat’s-eye yellow sample showed traces of calcium (0.068 wt.%
pearls,” we were granted permission to grind down one of CaO). All other measured elements were below the
the smaller samples until it was transparent. This section detection limit.
displayed a distinct layered structure perpendicular to the Microscopic examination of the dark green gem
direction of the fibrous prismatic calcite crystals (figure 30, showed a colorless seed plate with fine particles on each
right), as expected for shell. side (figure 33) and parallel green banding. The other gems
Having confirmed that these “cat’s-eye pearls” were were very clean, with only the light blue one showing
sculpted from calcitic shell, the question remained: reflective breadcrumb-like particles. No growth or color
Which mollusk could form a dark brown to black shell zoning was observed in those stones. Refractive indices
thick enough to cut a 10 mm pearl imitation? After some were 1.54–1.55, consistent with quartz, and Raman analy-
research, we found a private collection with a few mas- sis of the light blue and the yellow samples confirmed this
sive shells from Atrina vexillum (a pen shell) that were identification. We therefore identified all four samples as
thick enough to cut pearl imitations such as the ones synthetic quartz. While synthetic citrine is well known,
described in this report. dark green synthetic quartz is less common, and the aqua-
Thomas Hainschwang marine-like color varieties reported here are not often seen
( in the market.
GGTL Gemlab–GemTechLab Lore Kiefert (
Balzers, Liechtenstein Gübelin Gem Lab, Lucerne, Switzerland


Figure 31. These large
faceted samples, submit-
ted as amethyst (even
though dark green),
aquamarine, and citrine,
proved to consist of syn-
thetic quartz. The largest
weighs 162 ct. Photo by
Evelyne Murer, ©
Gübelin Gem Lab.

TREATMENTS acid process similar to that used for matrix opal from
Andamooka, Australia.
Sugar-acid treatment of opal from Wollo, Ethiopia. Several
For this experiment, we chose 12 mostly low-grade
gemological laboratories have recently identified “black”
opals from Ethiopia’s Wollo Province as smoke treated
Opal.pdf). The effectiveness of the smoke treatment is Figure 33. A close-up view of the dark green sample
probably related to the hydrophane character that is com- shows a seed plate (defined by arrows) and numerous
monly shown by Wollo opal. The porosity allows the fine particles. Photomicrograph by Lore Kiefert,
smoke to penetrate the opal structure deep enough to cre- © Gübelin Gem Lab; magnified 20×.
ate a dark bodycolor. Accordingly, one of us (FM) investi-
gated the possibility of treating Wollo opal using a sugar-

Figure 32. This FTIR spectrum of the synthetic citrine

is representative of all the synthetic quartz samples.
It shows a cutoff at ~3600 cm −1 and a small band at
5196 cm −1.



7000 6000 5000 4000 3000



Figure 34. These opals from Wollo, Ethiopia, are shown before (left) and after (right) sugar-acid treatment.
Darker bodycolors were produced in samples with a greater hydrophane character. The broken opals (center-
right and bottom-left) show the shallow penetration depth of the treatment. The samples measure from 13 × 9
× 7 mm (lower right) to 29 × 16 × 10 mm (upper left). Photos by F. Mazzero.

opals tumbled as irregular pebbles, with a white to yellow- land. More than 70 gemologists from 31 countries gath-
ish white bodycolor typical for opals from Wollo (figure ered to discuss developments in the field. The organizing
34, left). The samples were first heated at 90°C for 2 hours committee was led by Dr. Michael Krzemnicki, in collab-
in a solution containing 25% sugar by weight. Next they oration with colleagues at the Swiss Gemmological Insti-
were heated at 100°C for 3 hours in a 60% solution of tute SSEF, the Swiss Gemmological Society, George and
hydrochloric acid. All the opals turned a darker color (fig- Anne Bosshart, and Dr. Henry A. Hänni. The conference
ure 34, right). Some acquired a homogeneous, opaque, featured 12 sessions on topics ranging from colored stones
black bodycolor, while others darkened unevenly from to pearls and diamonds, analytical methods and gem treat-
grayish brown to gray. The play-of-color became more ments, and special sessions on Canadian gems, rare
intense in some samples and less vivid in others. Two of stones, and organic materials. The 48 talks and 14 interac-
the opals were broken open, revealing that the dark col- tive poster presentations covered a wide range of topics
oration penetrated only a few millimeters into the stones. and regions. The conference proceedings and excursion
The samples with the greatest hydrophane character (as guides are available at; some of the pre-
indicated by their ability to stick to the tongue) showed sentations are summarized below.
the darkest colors after treatment. Conversely, the more Dr. Thomas Armbruster (University of Bern, Switzer-
transparent and least hydrophane-like opals were least land) delivered the opening keynote address on gemology’s
affected by the treatment (e.g., inner portion of the upper- position at the interface of mineralogy and crystallography.
right sample in figure 34). As expected, the hydrophane Using the beryl group as an example, he demonstrated the
character appears to have facilitated the penetration of the similarities and differences in the crystal structures within
sugar and acid solutions into the opal. this group of minerals. Dr. Karl Schmetzer (Petershausen,
Even darker coloration in hydrophane opal may be Germany) described chemical zoning in trapiche tourma-
attainable by varying the carbon source and its concentra- line from Zambia, which is characterized by a strong nega-
tion, the nature of the acid and its concentration, and final- tive correlation between Ca and Na. Dr. Jürgen Schnellrath
ly the temperature and duration of heating in both solu- (Centro de Tecnologia Mineral, Rio de Janeiro) discussed
tions. Such experiments are in progress, and the results will unusual fiber distribution patterns in Brazilian cat’s-eye
give gemologists a better idea of what to expect for future quartz. Dr. Shang-i Liu (Hong Kong Institute of Gem-
treatments of this prolific type of opal. mology) presented results of a study on Cs- and Li-rich
Benjamin Rondeau ( beryl from Madagascar, using electron microprobe, LA-
Emmanuel Fritsch ICP-MS, FTIR, Raman analysis, and electron paramagnetic
Francesco Mazzero resonance spectroscopy.
Opalinda, Paris The pearl session included presentations by Dr.
Michael S. Krzemnicki on formation models for Tokki
Jean-Pierre Gauthier
cultured pearls, which form as attachments to larger bead-
Centre de Recherches Gemmologiques, Nantes, France
ed cultured pearls; Nick Sturman (GIA, Bangkok) on sepa-
rating natural from cultured Queen conch pearls (Strom-
CONFERENCE REPORTS bus gigas); and Federico Bärlocher (Farlang, Cernobbio,
32nd International Gemmological Conference. The bien- Italy) on the production and trade of Melo pearls from
nial IGC was held July 13–17, 2011, in Interlaken, Switzer- Myanmar.


Walter Balmer (Department of Geology, Chulalong- the predominant buyers, and how these global influences
korn University, Bangkok) addressed the characterization are pushing some diamond and gem prices to new heights.
of rubies from marble deposits in Myanmar (Mogok), A round-up of the auction market, including information
Vietnam (Luc Yen) and Tanzania (Morogoro and Uluguru on commissions and insurance, as well as recommenda-
Mountains) using UV-Vis spectroscopy. John Koivula tions on how to best work in the auction world, was pre-
(GIA, Carlsbad) showed how high-temperature fusion of sented by Stephen Whittaker of Fellows & Sons
corundum mimics residues in heat-treated rubies and sap- (Birmingham, UK). He also explained the use and costs of
phires. Terry Coldham (Sapphex, Sydney) advocated technology in the auction business.
greater cooperation between researchers and producers in The focal point of the conference was introducing a
shedding light on the genesis of basaltic sapphires. new valuation qualification, CAT (Certificate of Appraisal
Edward Boehm (RareSource, Chattanooga, Tennessee) Theory), which was presented by Heather McPherson, a
offered insight into current colored stone market trends, Fellow of the NAG Institute of Registered Valuers (Coal-
and Brad Wilson (Alpine Gems, Kingston, Canada) ville, UK). This training program is intended to replace the
described a wide variety of recent finds on Baffin Island, JET (Jewellery Education & Training) valuation course and
Canada. will be launched by spring 2012 as a new requirement to
Dr. Thomas Pettke (University of Bern) gave the con- become an Institute Registered Valuer.
ference’s second keynote talk, in which he reviewed LA- Harry Levy, president of the London Diamond Bourse,
ICP-MS applications and offered an outlook on the poten- explained the valuation of diamonds and colored stones,
tial of this method in gemology. The poster sessions gave with information on nomenclature, treatments, and ori-
insights into current research, including presentations on gins. The importance of treatment disclosure was the
jadeite trading in China (Elizabeth Su, DuaSun Collection, main theme of his presentation. Steve Bennett of GemsTV
Shanghai) and rare-earth coloration in bastnäsite (Dr. Franz (Worcestershire, UK) described how education, training,
Herzog, Swiss Gemmological Institute SSEF, Basel). and vertical integration have been the key factors in his
The conference was preceded by excursions to the family-run business.
Natural History Museum in Bern and mineral localities in Mehdi Saadian (
the Ticino Alps. A three-day trip to the Binn Valley and GIA, London
Zermatt immediately followed the conference. Both tours
took participants to areas of mineralogical and geological
interest in Switzerland. Adopting Dr. Nguyen Ngoc Khoi’s ERRATA
proposal, the delegates unanimously decided to hold the
1. For figure 1 of the Fall 2011 CVD synthetic diamond
2013 IGC in Vietnam.
article by B. Willems et al. (pp. 202–207), the correct
Laurent Cartier (
photographer of the images in this composite photo
Swiss Gemmological Institute SSEF, Basel
was Jian Xin (Jae) Liao.
2011 NAG Institute of Registered Valuers Conference. 2. Figure 18 of the Fall 2011 Vietnamese corundum article
The National Association of Goldsmiths IRV 23rd annual by N. N. Khoi et al. (pp. 182–195) shows the absorption
conference took place September 24–26 at its usual venue, spectrum of a sapphire, rather than a ruby. G&G thanks
Loughborough University, Leicestershire, UK. Approxi- Nathan Renfro for this correction.
mately 150 delegates attended the various presentations, 3. The Fall 2011 GNI entry by D. Beaton (pp. 247–248)
including hands-on workshop sessions. should have indicated that the strong absorption peaks
Market trends and industry issues were discussed by at 376 and 386 nm were attributed to Fe 3+ pairs (376
Richard Drucker of Gemworld International (Glenview, nm) and Fe3+ (386 nm). G&G apologizes for introducing
Illinois). He explained how the market is changing locally errors in the peak assignments, and thanks Nathan
and internationally with China and India now becoming Renfro for the correction.


Susan B. Johnson
Jana E. Miyahira-Smith

The Diamond Compendium gy for airborne gravity gradient sur- role in the corridors of luxury—and in
veying are out of place and should be civil conflict. But Ms. Oltuski, whose
By DeeDee Cunningham, 888 pp.,
reworked to avoid confusion with father and grandfather were both dia-
illus., publ. by NAG Press, London,
magnetic surveying. Apart from these mond dealers, understands the book
2011. US$225.00
items, the information is well editors’ maxim and manages to cap-
This hardcover tome of nearly 900 researched and clearly written. ture all of these topics honestly and in
pages is filled with pertinent illustra- The author takes great care in pre- a deeply personal way to create an
tions. The book is comprised of 20 senting the correct terminology. She engaging, well-written work.
chapters divided into sections that explains the different meanings of Ms. Oltuski is a skilled writer
cover all aspects of diamonds: origin, form and habit in crystallography, and who can transform the everyday deal-
crystallography, global occurrences, why lonsdaleite (a hexagonal poly- ings of a diamond office into enter-
exploration, mining methods, cutting, morph of diamond) is not a kind of taining stories, without resorting to
polishing, grading, and identification. diamond but a different mineral. The the sensationalist tales of chasing
Simulants and synthetic diamonds chapters on evaluating carat weight, multimillion dollar stones and dodg-
are also reviewed. clarity, color, and cut offer useful tips ing international jewel thieves that
The compendium is not meant to on how to hold and angle the diamond other authors have created to pack
be read cover to cover; it is a reference for best results. The chapter on color drama into an essentially mundane
work to be consulted on specific top- examines how color is perceived, buy-and-sell world. The result is a
ics, some of which are explored in while others review color treatments realistic account of New York’s dia-
greater scientific depth. As expected and recutting to enhance brilliance. mond community, down to the duct-
from a highly skilled gemologist and The chapters on identification of natu- taped jewelry cases.
jeweler, the sections on crystallogra- ral diamond, simulants, and synthetic The early chapters focus on the art
phy, physical properties, cutting, pol- diamond are very good. of the deal. Stories are told of her
ishing, and grading are the strongest, The Diamond Compendium took father’s negotiations with clients
while the text on global occurrences seven years to compile, and the infor- (who’s fibbing and who’s really offer-
contains a few inaccuracies. For mation is up to date to 2007. The wide ing genuine prices), her mother’s stops
instance: range of topics makes it an important to help with stone deliveries on the
• The Lomonosov kimberlite cluster, reference source and handbook with way to her own job, the weight her
100 km north of Arkhangelsk in practical tips for use in gemology. grandparents’ reputation still carried
northwestern Russia, is not close to A. J. A. (BRAM) JANSE years later, and a childhood where
the Finnish border but 500 km to Archon Exploration security and secrecy came home from
the east of it. Carine, Western Australia the office.
• The Golconda alluvial diamond In the middle chapters she artfully
field is not located near the ancient weaves in personal anecdotes and
observations to transform the oft-told
Golconda fortress but at least 250 Precious Objects: A Story of clichés about New York’s diamond
km to the southeast.
Diamonds, Family, and a industry into fresh, appealing reading.
• No kimberlites or other primary The accounts of how diamonds are
diamondiferous rocks have ever
Way of Life
formed deep beneath the earth, the
been found in Guyana, and certain- By Alicia Oltuski, 370 pp., publ. by workings of De Beers’s sight system
ly not the 14 kimberlites quoted on Scribner, New York, 2011. US$24.00 (albeit a bit out of date in today’s
page 119. Book editors often say every story has multi-source environment), the mores
There is also a slight problem on been told before, so the secret of a of the Hasidic Jewish community, and
page 207, in the paragraph on airborne good book lies in the telling. Cer- 47th Street’s love-hate relationship
prospecting. Most airborne prospect- tainly, the past two decades have seen with Martin Rapaport’s price list—
ing is carried out to detect local differ- numerous books describing how dia- they are all here, but in a new dress.
ences in the earth’s magnetic field monds are formed deep within the Her portrait of her grandfather
caused by mineral deposits (such a earth, how De Beers gained control of Yankel (“Jack” to the New Yorkers) is
survey is shown in photo 5.1). The rough diamond production, how an intimate account of how diamond
two sentences about Falcon technolo- stones are bought and sold, and their families came into being after the


upheavals of World War II. Yankel mineral inclusions. The glossy text is MEDIA REVIEW
was a natural trader, scratching a liv- accompanied by photos and diagrams, What’s Hot in Tucson 2010. DVD
ing in post-war Europe by scrounging more than 300 references, and a help- (three discs), approx. 6 hours, released
scarce items and swapping them for ful index. The authors avoid confusing by Blue Cap Productions [
merchandise he could swap for even and excruciating detail and provide], Marina del
more goods. That led him to dia- references to relevant literature for Rey, CA, 2011. $39.95
monds. Yankel’s story is one of sur- further particulars.
What’s Hot in Tucson 2010—Gems
vival, like that of many New Yorkers Everything gets a mention. The
& Jewelry. DVD, approx. 2 hours,
from his generation who had come “Origin” chapter, for instance, dis-
released by Blue Cap Productions
there to begin life anew and raise a cusses cratons, diamonds from litho-
family. The stories of those who spheric mantle roots to upper parts of
Marina del Rey, CA, 2011. $24.99
experienced the horrors of war, the the lower mantle, types of diamond,
Holocaust, and mass displacements diamonds in metamorphic crustal What’s Hot in Tucson 2010 is the
cannot be told often enough before rocks from Kokchetav to Jack Hills, fourth in Blue Cap’s series of videos
they pass on. diamonds in impact structures and that take viewers behind the scenes of
In the end, Ms. Oltuski’s book is meteors, and presolar specimens. The the city’s major mineral shows. The
an honest look at the diamond trade. “Morphology” chapter recognizes three DVDs, cohosted by David
And in telling this story, it becomes three major types: monocrystalline, Wilber and Bob Jones, provide hours
obvious that the book’s real precious fibrous, and polycrystalline (including of interviews with mineral dealers at
objects are not the gemstones, but her types of carbonado). Habits are divid- five different venues: the InnSuites
family members and the other men ed into regular, irregular, twins, and hotel, the Pueblo Inn Gem & Mineral
and (few) women who inhabit New macles, and photos show representa- Show, the Fine Minerals International
York’s diamond district. tive crystals. In the “Colors” chapter house, the Westward Look resort, and
RUSSELL SHOR there are details on colorless, yellow, the TGMS Main Show at the Tucson
Gemological Institute of America blue, brown, pink, purple, green, and Convention Center.
Carlsbad, California brown-spotted diamonds. Well-cho- The program highlights the
sen photos in the “Surface Textures” newest and finest minerals in the mar-
chapter show various surface features ket, showing superb examples of exot-
occurring preferentially on certain ic and classic mineral specimens. The
Diamonds in Nature: crystal faces, such as terraces, hill- cohosts are subject experts, and the
A Guide to Rough Diamonds ocks, micro-disks and micro-pits, interviews provide valuable informa-
frosting, ruts, edge abrasion, scratch- tion about the specimens, such as
By Ralf and Michelle C. Tappert, locality, formation, and mining tech-
es, and percussion marks. The “Inclu-
142 pp., illus., publ. by Springer niques. The production values are
sions” chapter recognizes three types
Verlag, Heidelberg, Germany, 2011. high, allowing the viewer to appreci-
by origin (protogenetic, syngenetic,
US$60.00 ate the minerals’ beauty and details.
and epigenetic) and four types by min-
Husband-and-wife team Ralf and eralogy (peridotitic and eclogitic This video series also includes the
Michelle Tappert offer a prime exam- suites as lithospheric inclusions, an first installment of What’s Hot in
ple of how a scientific text should be asthenosphere/transition zone suite, Tucson—Gems & Jewelry, which
written and illustrated. The subject and a lower mantle suite as sublitho- offers an insider’s look at the Ameri-
matter is handled in clear, concise spheric inclusions). can Gem Trade Association show at
sentences accompanied by Mr. The volume also contains a list of the Tucson Convention Center. The
Tappert’s superb photographs. As it unconfirmed and rare mineral inclu- AGTA GemFair is where many of the
says in the foreword, there are many sions, with a selected reference for year’s trends and new products debut.
books that discuss diamond’s optical, each. Mineral inclusions of all types This segment is hosted by Delphine
physical, and gemological properties, are covered in the text, accompanied Leblanc, who conducts interviews on
but few on rough diamonds and how by beautiful color illustrations. Car- the floor with some of the biggest
they appear in nature. Diamonds in bon isotopes and the sources of car- names at the show, including cutters,
Nature does exactly this and does it bon (organic, mantle, or carbonates) designers, and dealers in estate jewel-
admirably well. are discussed briefly, as are the ages of ry, specialty colored gemstones,
This book is not for the general diamonds. pearls, and colored diamonds.
reader but for diamond gemologists In summary, I recommend this From novice to expert, there is
and geologists. It gives a comprehen- thoroughly informative text to all something for everyone in this eight-
sive account of every feature of rough readers interested in the subject of hour DVD package.
diamonds, including their origin, mor- rough diamonds. MICHELE KELLEY
phology, colors, surface textures, and A. J. A. (BRAM) JANSE Monmouth Beach, New Jersey


Brendan M. Laurs
Geographical origin classification of gem corundum using
elemental fingerprint analysis by laser ablation induc-
REVIEW BOARD tively coupled plasma mass spectrometry. M.-M.
Pornwilard, R. Hansawek, J. Shiowatana, and A.
Edward R. Blomgren
Owl’s Head, New York
Siripinyanond [], International
Journal of Mass Spectrometry, Vol. 306, No. 1, 2011, pp.
Annette Buckley 57–62,
Austin, Texas
Various market sectors are increasingly asking for origin identi-
Jo Ellen Cole fication of rubies and sapphires. Because the chemical composi-
Vista, California tion of corundum reflects its geographic origin, fingerprinting
R. A. Howie techniques based on trace-element analysis are of great inter-
Royal Holloway, University of London est. Several methods have been used—particle-induced X-ray
emission/proton-induced gamma-ray emission (PIXE/PIGE),
Edward Johnson
micro-PIXE, laser-induced breakdown spectrometry (LIBS), and
GIA, London
energy-dispersive X-ray fluorescence (EDXRF) analysis—but
Michele Kelley these techniques are either time-consuming or provide insuffi-
Monmouth Beach, New Jersey cient detection limits.
Guy Lalous This study employed laser ablation–inductively coupled
Academy for Mineralogy, Antwerp, Belgium plasma–mass spectrometry (LA-ICP-MS), combined with mul-
tivariate data analysis, to classify the origin of 58 gem corun-
Kyaw Soe Moe
dum samples. Trace-element concentrations were analyzed by
GIA, New York
LA-ICP-MS to construct a data matrix, with columns repre-
Keith A. Mychaluk senting the concentration of 10 elements (B, Si, Zn, Ga, Sn,
Calgary, Alberta, Canada Mg, Ti, V, Cr, and Fe) in each stone. Two multivariate statisti-
Joshua Sheby cal techniques—principal-component analysis from factor
New York, New York analysis and linear discriminant analysis (LDA)—were
employed to differentiate and classify the samples.
James E. Shigley
This approach identified the geographic origin of both
GIA, Carlsbad
homogenously colored and multicolored corundum samples
Russell Shor from six countries. LDA provided good differentiation between
GIA, Carlsbad rubies from Southeast Asia and Africa, as well as blue sapphires
Jennifer Stone-Sundberg
Portland, Oregon
Rolf Tatje
Duisburg, Germany This section is designed to provide as complete a record as
practical of the recent literature on gems and gemology. Articles
Dennis A. Zwigart are selected for abstracting solely at the discretion of the section
State College, Pennsylvania editors and their abstractors, and space limitations may require
that we include only those articles that we feel will be of greatest
interest to our readership.
Requests for reprints of articles abstracted must be addressed to
the author or publisher of the original material.
The abstractor of each article is identified by his or her initials at
the end of each abstract. Guest abstractors are identified by their
full names. Opinions expressed in an abstract belong to the abstrac-
tor and in no way reflect the position of Gems & Gemology or GIA.
© 2011 Gemological Institute of America


from Madagascar and Nigeria. LDA mapping, with the use appendices in the CITES regulations organize and list the
of a normalization factor, identified the origin of variously species. Appendix I covers species that are greatly threat-
colored corundum with 80% accuracy. ERB ened with extinction. Materials from these species (e.g.,
tortoise shell) can only be traded for exceptional purposes,
Nomenclature of the tourmaline-supergroup minerals. D. such as scientific research. Appendix II species are not nec-
J. Henry, M. Novak, F. C. Hawthorne, A. Ertl, B. L. essarily facing extinction, but their trade must be con-
Dutrow, P. Uher, and F. Pezzotta, American Miner- trolled to avoid threatening their survival. Elephant ivory is
alogist, Vol. 96, 2011, pp. 895–913, an example of an Appendix II material, and some argue that
10.2138/am.2011.3636. it should be listed in Appendix I. To meet CITES require-
Tourmaline is a chemically complex mineral group that ments for legal trade, Appendix II species require a certifi-
presents challenges for both chemical analysis and miner- cate issued by an authority of the country or state of
alogical classification. Tourmalines are typically not differ- export/import. Species listed in Appendix III are protected
entiated according to mineral species by gemologists in at least one country that has requested other CITES par-
because they cannot be distinguished by normal gem test- ties’ assistance in controlling the trade of that species.
ing methods (and in many cases, gem tourmalines are As countries become more familiar with these regula-
elbaites). This mineral supergroup is comprised of 18 tions, many gemological laboratories are seeing an
species currently recognized by the International Miner- increase in organic gems submitted to determine if they
alogical Association, represented by the generalized struc- are protected. The distinction can sometimes be made just
tural formula XY3Z6(T6O18)(BO3)3V3W. The most common by using magnification. The presence of Schreger (also
constituents of each site are: known as “engine turn”) lines indicates either elephant or
mammoth ivory. Mammoth ivory is not protected by
X = Na+, Ca2+, K+, vacancy
CITES because the animal is already extinct, and the angle
Y = Fe2+, Mg2+, Mn2+, Al3+, Li+, Fe3+, Cr3+
at which the Schreger lines intersect can be diagnostic for
Z = Al3+, Fe3+, Mg2+, Cr3+
separating it from elephant ivory. The lines are not always
T = Si4+, Al3+, B3+
apparent under magnification, and more sophisticated (and
B = B3+
expensive) instruments such as X-ray computed microto-
V = OH–, O2–
mography can be used for more challenging ivory/mam-
W = OH–, F–, O2–
moth separations. Similar distinctions are needed between
The tourmaline supergroup can be divided into several Corallium vs. Stylaster red or pink corals. Growth struc-
groups and subgroups. Alkali, calcic, and X-vacant primary tures observed with magnification are diagnostic, but
groups are recognized on the basis of their X-site occupan- these are not always visible. Raman spectroscopy is useful
cy. Because cations with different valence states can occu- for making this separation.
py the same site, coupled substitutions of particular cations A large percentage of the protected species currently
are required to relate the compositions of the groups. used as gem materials can be easily identified by well-
Within a relevant site, the dominant ion of the dominant equipped gemological laboratories. The addition of other
valence state is used as the basis of nomenclature. species to the CITES appendices will require the use of
The article describes the major tourmaline groups and more sophisticated nondestructive testing methods to sep-
the recognized and prospective mineral species they arate them. JEC
encompass. It defines the concepts central to tourmaline
classification and provides a hierarchical approach to The role of matrix proteins in the control of nacreous layer
determining species nomenclature according to the deposition during pearl formation. X. J. Liu, J. Li, L.
amount of information available on a specimen’s chemical Xiang, J. Sun, G. L. Zheng, G. Y. Zhang, H. Z. Wang,
composition. Finally, the authors discuss each of the rec- L. P. Xie, and R. Q. Zhang, Proceedings of the Royal
ognized tourmaline species, providing lists of color-based Society B, 2011, pp. 1–8,
varietal names as well as obsolete, discredited, or misiden- rspb.2011.1661.
tified species. JES Formed by a biomineralization process, pearl nacre is com-
posed of aragonite platelets and biological macromolecules.
Organic gems protected by CITES. S. Karampelas and L. The platelets are arranged in continuous parallel layers sep-
Kiefert, InColor, No. 15, Fall/Winter 2010, pp. 20–23. arated by interlamellar organic sheets and intercrystalline
Organic gems, defined as those formed through biological organic membranes. This study focused on the deposition
processes, are often the product of endangered animal or of the platelets on pearl nuclei, and the gene expression of
plant substances. Among numerous conventions and five nacre matrix proteins in the pearl sac during early
treaties governing the local, national, and international stages of formation. The authors found that pearls begin
trade in endangered species materials, the most widely rec- forming with irregular CaCO3 deposition on the nucleus,
ognized is the Convention on International Trade in followed by further deposition that becomes more and
Endangered Species of Wild Fauna and Flora (CITES). Three more regular until the mature nacreous layer has formed.


The low-level expression of matrix proteins during the broad commercial market because it combines beauty,
irregular CaCO3 deposition suggests that the process may durability, rarity, a variety of sizes and shapes, and an
not be initially controlled by the organic matrix. Instead, attractive range of price points.
significant expression of matrix proteins in the pearl sac The authors provide a list of tourmaline colors, causes
was detected 30–35 days after nucleus implantation. This of color, varietal names, and probable species. Due to tour-
was accompanied by a change in aragonite deposition from maline’s pronounced and variable dichroism, optimal color
large, irregular crystals to smaller, ordered crystals. This is achieved during the cutting process. Further influencing
ordered deposition was controlled by the organic matrix. color is natural or laboratory irradiation of some pale tour-
These results suggest that the bioactivities of matrix malines, which commonly produces vivid pink and red
proteins are critical for nacre formation and pearl develop- stones, and heat treatment of dark blue or green stones to
ment, by controlling the shape, size, nucleation, and lighten or intensify their color. Interestingly, gem tourma-
aggregation of CaCO3 crystals. JES line has not yet been synthesized, probably due to its
chemical complexity.
Study of structural and valence state of Cr and Fe in Copper-bearing gem tourmaline was discovered in the
chrysoberyl and alexandrite with EPR and Möss- 1980s in the Brazilian state of Paraíba, and has become one
bauer spectroscopy. V. S. Urasov, N. A. Gromalova, of the world’s most desirable gems, largely due to its vivid
S. V. Vyatkin, V. S. Rusakov, V. V. Maltsev, and N. “neon” blue-to-green and purple hues. The article
N. Eremin, Moscow University Geology Bulletin, overviews the geographic locales, coloring mechanisms,
Vol. 66, No. 2, 2011, pp. 102–107. and absorption spectra of this tourmaline variety, which
The authors investigated the valence and structural distri- can rival diamonds in cost.
butions of chromium and iron in chrysoberyl. The sam- Gem tourmalines form in magmatic, pegmatitic envi-
ples consisted of alexandrite from the Ural Mountains and ronments and in metamorphic rock, with variable geolog-
an unknown deposit, chrysoberyl from Tanzania, and syn- ical ages—as recent as 6 million years to between 100 and
thetic alexandrite produced by Czochralski, flux, and 550 million years ago. Inclusions mainly consist of
hydrothermal methods. EPR and Mössbauer spectroscopy trapped fluids, randomly arranged capillary-like tubes,
were used to determine the valence states, and limited partially healed fractures, and occasional growth tubes.
annealing studies in air and argon were performed to deter- The article touches on the gem’s distinguished and col-
mine their impact on structural distributions. orful history, including its role in the jewels of royalty. Of
The two Al 3+ sites in the chrysoberyl structure, M1 particular note are a huge “ruby” (actually red tourmaline)
and M2, are slightly distorted octahedral voids with differ- in the great Czech crown of Saint Wenceslas, the specimens
ing symmetries and volumes. The authors cite earlier in the Kremlin’s Treasure Room, and the Dowager Empress
studies to establish that Fe 3+ cations substitute for Al3+ Tz’u Hsi’s passion for carved pink tourmaline. ERB
ions in the (larger) M2 sites exclusively. The present work
On the words used as names for ruby and sapphire. L. A.
shows that while Fe2+ enters the chrysoberyl structure,
Lytvynov, Functional Materials, Vol. 18, No. 2,
only some Fe3+ enters the structure and the balance exists
2011, pp. 274–277.
in other phases such as Fe2O3. The wide range of Fe2+ to
Fe 3+ ratios in the various samples indicates crystal forma- This article examines the trade names of ruby and sap-
tion under a variety of redox conditions. phire from different cultures and historical periods. In fact,
From the annealing studies, the authors suggest that many names that contain the noun “ruby” preceded by an
the proportion of Cr 3+ occupancy between the M1 and M2 adjective are not rubies at all, but other varieties of miner-
sites is dependent in part on the crystallization tempera- als. The author gives the etymology of many of these trade
ture. At higher temperatures such as those achieved with names and a comprehensive chart. JS
synthetic growth techniques or annealing, the ratio of M2
to M1 occupancy increases. One possible rationale the
authors propose is thermal diffusion of Cr 3+ from the M1 DIAMONDS
site to the M2 site. JS-S Bistable N2-H complexes: The first proposed structure of a
H-related colour-causing defect in diamond. J. P.
Tourmaline: The kaleidoscopic gemstone. F. Pezzotta Goss, C. P. Ewels, P. R. Briddon, and E. Fritsch,
[] and B. M. Laurs, Elements, Diamond and Related Materials, Vol. 20, No. 7,
Vol. 7, No. 5, 2011, pp. 333–338, 2011, pp. 896 –901,
10.2113/gselements.7.5.333. j.diamond.2011.05.004.
This article examines one of the most popular and spectac- For some time, scientists have thought that hydrogen-relat-
ular colored stones (and this abstractor’s favorite gem). ed defects in diamond, detectable by spectroscopy, might
Tourmalines are found in virtually every color of the rain- contribute to their coloration, but the exact nature of these
bow, including complex and beautifully zoned and occa- defects has remained unclear. The hydrogen content of dia-
sionally chatoyant crystals. The gem is well suited for a monds varies widely. In this article, the authors propose a


model for a hydrogenated substitutional nitrogen pair tures. Based on a study of 330 gem-quality crystals from
defect (N2-H), which they predict will exhibit optical the Ekati mine in the Northwest Territories of Canada,
absorption above 600 nm and thus contribute to diamond the authors developed a methodology to distinguish
coloration. They suggest that this defect is one of a family between dissolution events that took place in the mantle
of nitrogen-hydrogen defects, which due to covalent nitro- versus those occurring in the ascending kimberlitic
gen-hydrogen bonding might explain several absorptions magma. The authors inferred that the majority of the sam-
observed in the spectra of diamonds. This defect could exist ples (~72%) exhibited resorption features produced within
in both neutral and negative charge states. Structural differ- kimberlite. Approximately 13% experienced resorption in
ences associated with these states may also contribute to the mantle, while the remainder displayed resorption fea-
the thermochromic and photochromic behavior exhibited tures produced in both environments.
by so-called chameleon diamonds. JES The diamonds with mantle-generated resorption fea-
tures had a different thermal history, as evidenced by their
Deep mantle cycling of oceanic crust: Evidence from dia- lower nitrogen content and greater degree of nitrogen aggre-
monds and their mineral inclusions. M. J. Walter, S. gation. The diamonds with kimberlite-induced resorption
C. Kohn, D. Araujo, G. P. Bulanova, C. B. Smith, E. could be subdivided into two categories: resorption in the
Gaillou, J. Wang, A. Steele, and S. B. Shirey, presence of either an aqueous fluid or kimberlitic magma.
Sciencexpress, Vol. 334, No. 6052, 2011, pp. 54–57, The authors provide photos of diamond crystals that illus- trate the various resorption styles. They suggest that dia-
In plate tectonics, basaltic oceanic crust subducts on monds from kimberlite pipes (even those separated by geo-
lithospheric slabs into the mantle. Seismological studies logic space or time) would be expected to have similar man-
extend this subduction process down into the lower man- tle-generated resorption, but might display different magma-
tle (>600 km depth), and geochemical observations indi- generated resorption features depending on the retention or
cate the return of some subducted oceanic crustal materi- loss of aqueous fluid from the ascending magma. JES
al to the upper mantle (~200–500 km depth) in convective
mantle plumes. On the basis of their mineral inclusions, Nano and sub-micro inclusions as probes into the origin
geologists believe that most diamonds found at the and history of natural diamonds. J. Purushothaman,
earth’s surface originated in the lithosphere at depths of P. R. Sajanlal, M. Ponnavaikko, and T. Pradeep,
less than 200 km. Diamond and Related Materials, Vol. 20, No. 7,
In this article, the authors analyzed mineral inclusions 2011, pp. 1050–1055,
in “superdeep” type IIa diamonds from the Juina-5 kim- j.diamond.2011.06.003.
berlite in Mato Grosso, Brazil. These unusual diamonds Of five diamonds examined, one contained a group of
host inclusions with compositions comprising the entire micro-inclusions. Non-diamond Raman peaks at 514,
assemblage of mineral phases expected to crystallize from 510, and 490 cm –1 were detected in different areas of
basalt in the lower mantle. Mineralogical evidence indi- the stone, suggesting a group of micro-inclusions with
cates that originally homogeneous silicate phases (trapped various constituents. This was confirmed by a Raman
in the diamonds as inclusions during growth) exsolved intensity image, which showed bright and dark zones
into composite mineral assemblages. This unmixing is in the range from 750 to 100 cm –1. Energy-dispersive
interpreted as having occurred at a later time during the chemical analysis gave the composition of a pyroxene
convective ascent of the diamonds from the lower to the (Fe, Mg, and Ca silicate) with a ternary perovskite
upper mantle. These diamonds record a history of upward structure. It also detected uneven distributions of Cr,
transport on the order of 500–1,000 km in the mantle, Ru, Nb, Co, and Ni throughout the micro-inclusion
long before being sampled by kimberlite magmas and assemblage. Fe and Cr were clearly observed in mass
brought to the surface. Since the carbon isotopic signature spectra using laser desorption ionization–mass spec-
of these superdeep diamonds is consistent with a surface- trometry (LDI-MS).
derived source of carbon, the authors conclude that the The authors concluded that this diamond was of
deep carbon cycle extends all the way from the crust to peridotitic origin, as evidenced by the presence of ferro-
the lower mantle. JES magnesian pyroxene minerals and Cr, and the absence
of aluminum, alkalis, and rare-earth elements. This
Diamond resorption: Link to metasomatic events in the nondestructive micro-inclusions study showed poten-
mantle or record of magmatic fluid in kimberlitic tial for defining the genesis of not only diamonds but
magma? Y. Fedortchouk and Z. H. Zhang, Canadian also of other gem materials. KSM
Mineralogist, Vol. 49, No. 3, 2011, pp. 707–719, Seismic architecture of the Archean North American
Because of their lengthy and complex history of growth mantle and its relationship to diamondiferous kim-
and dissolution in the earth, diamond crystals display a berlite fields. S. Faure, S. Godey, F. Fallara, and S.
tremendous variation in their morphology and surface fea- Trépanier, Economic Geology, Vol. 106, No. 2,


2011, pp. 223–240, deposits in other parts of the world, these diamonds appar-
geo.106.2.223. ently did not originate from the subcratonic lithosphere.
The architecture of the base of the lithosphere beneath Approximately 3,000 diamonds were characterized for
cratons in Canada and the United States has been defined this study. Raman spectroscopy of mineral inclusions indi-
for the first time in three dimensions using a high-resolu- cates the diamonds formed under high-pressure conditions
tion, Rayleigh wave-phase-velocity, tomographic model of in the lower mantle at the termination of subduction of
the upper mantle (30–250 km depth). There is remarkable oceanic crust by continental collision. Using IR spec-
agreement between data obtained from mantle xenoliths troscopy, the authors distinguished two groups of diamonds
and the lithospheric base obtained from this geophysical according to their nitrogen abundance and aggregation state.
model. The inferred base also corresponds to the petrologic They suggest that these two groups originated from differ-
definition of the boundary between the lithosphere and ent mantle sources and propose the following delivery
the asthenosphere. mechanism. After subduction, the eclogitic host rocks were
An important conclusion of this study is that diamondif- stranded for a period in the lower mantle under high-pres-
erous kimberlites are not located over the deepest parts of sure conditions where diamond formation took place, and
the cratonic roots (~180–240 km). Rather, they seem to occur were then exhumed to the upper mantle/lower crust. At
most often over the sloping edges of the lithosphere-astheno- some later time, diamonds (along with garnet and pyroxene)
sphere boundary that surround the deepest part of the craton- were transported to the surface by alkali basaltic intrusions.
ic root zone. The starting point for kimberlite magma ascen- The diamonds exhibit evidence of pervasive deforma-
sion to the surface originates from these areas at depths of tion during growth, resulting in crystal imperfections,
160–200 km. This depth interval delineates the region of the strongly aggregated nitrogen, a reduction of the second-order
upper mantle where diamonds should have been stable since Raman peak that normally would be more intense, and
the Archean, and over which all known existing and poten- exceptional durability in industrial applications. JES
tial future diamond mines in Canada are located. JES
Start of the Wilson cycle at 3 Ga shown by diamonds from
A secondary ion mass spectrometry (SIMS) re-evaluation the subcontinental mantle. S. B. Shirey and S. H.
of B and Li isotopic compositions of Cu-bearing Richardson, Science, Vol. 333, No. 6041, 2011, pp.
elbaite from three global localities. T. Ludwig, H. R. 434–436,
Marschall, P. A. E. Pogge von Strandmann, B. M. The Wilson cycle describes the set of major plate tecton-
Shabaga, M. Fayek, and F. C. Hawthorne, Miner- ic processes whereby continental plates were dispersed
alogical Magazine, Vol. 75, No. 4, 2011, pp. and then reassembled on the earth’s surface. These pro-
2485–2494. cesses are associated with crustal growth, mountain
Copper-bearing elbaites are recovered from pegmatite-relat- building, and the occurrence of ore deposits. This multi-
ed deposits in Brazil, Mozambique, and Nigeria. Chemical stage cycle has repeated itself throughout geologic time,
analyses of 10 samples from these three geographic sources and its operation since the Precambrian is apparent in
using the SIMS technique revealed distinct differences in the geologic record of crustal rocks. The timing of the
boron and lithium isotope compositions, which offer valu- start of this cycle has been uncertain, however. Mineral
able tourmaline provenance information. The isotopic com- inclusions, encapsulated in diamonds billions of years
positions for these two elements are relatively homogeneous ago during their formation in the subcontinental litho-
even across chemical- and color-zoned samples. The two iso- sphere, may provide information on this fundamental
topes are also within the range of previously published data geologic event.
for granitic and pegmatitic tourmalines. JES The authors analyzed existing data on 4,287 silicate
and 112 sulfide inclusions in diamonds recovered from
Spectroscopic research on ultrahigh pressure (UHP) ancient cratons in Australia, southern Africa, and Siberia.
macrodiamond at Copeton and Bingara NSW, east- These inclusions were classified, on the basis of their
ern Australia. L. Barron, T. P. Mernagh, B. J. Barron, chemical composition, as either peridotitic or eclogitic.
and R. Pogson, Spectrochimica Acta A, Vol. 80, No. Only peridotitic inclusions formed prior to 3.2 billion
1, 2011, pp. 112–118, years ago (Ga), while diamonds with eclogitic inclusions
j.saa.2011.03.003. became more prevalent after 3.0 Ga. The researchers con-
Since the 19th century, an estimated one million carats of cluded that the change in inclusion mineralogy at this
diamonds (average weight ~0.25 ct) have been mined from time resulted from the capture of eclogite and diamond-
numerous secondary placer deposits along the eastern coast forming fluids in the subcontinental mantle due to sub-
of Australia, the vast majority from the Copeton-Bingara duction and continental collision associated with the
area of New South Wales. Most were recovered between onset of plate tectonics. The start of the Wilson cycle at
1867 and 1922. The absence of typical indicator minerals in 3.0 Ga marked a change from crustal modification by recy-
these sedimentary deposits has hampered geologists’ efforts cling and other near-surface processes to plate tectonics, a
to identify the primary sources of the diamonds. Unlike shift that has continued to the present day. JES


GEM LOCALITIES Brazil’s emerald mines are located in the states of Minas
Gerais, Bahia, Goiás, and Tocantins. Most deposits of eco-
Different origins of Thai area sapphire and ruby, derived
nomic interest are hosted by mica schists. The emeralds’
from mineral inclusions and co-existing minerals.
gemological properties directly reflect the geological-min-
S. Saminpanya and F. L. Sutherland, European
eralogical environment in which they formed. The geolog-
Journal of Mineralogy, Vol. 23, No. 4, 2011, pp.
ic conditions in the various Brazilian deposits have many
similarities, so the emeralds’ properties may also overlap.
Colombian emeralds have distinctive three-phase inclu-
The authors used SEM-EDS analysis to identify and mea- sions and growth structures, while mica inclusions
sure the chemical composition of mineral inclusions in (biotite-phlogopite) are the most common solid inclusions
gem corundum from Thailand. The mineral phases in the found in Brazilian stones. Growth phenomena are normal-
sapphires included alkali feldspar (sanidine), nepheline, zir- ly not well developed in Brazilian emeralds.
con, and spinel (hercynite and magnetite-hercynite). Those Chemical analyses of Brazilian and Colombian speci-
identified in rubies included pyrope, diopside (fassaite), and mens were acquired by LA-ICP-MS, and several elements
sapphirine. (Cr, V, Fe, Na, Mg, Ga, and Cs) are discussed. The V/Cr
Based on the composition of these inclusions, the ratio provides a reliable indication of the origin of color.
authors conclude that Thai rubies crystallized in the upper The absorption spectra of Colombian emeralds normally
mantle in high-pressure metamorphic rocks of ultramafic show variable contributions from both Cr 3+ and V3+.
or mafic composition (they suggest a garnet pyroxenite or Colombian specimens contain very little iron and gener-
granulite). The sapphires had higher Ga, Ti, Fe, Ta, Nb, ally do not contain bands related to Fe3+. Brazilian emer-
and Sn contents than the rubies. The authors suggest that alds have “combination spectra” composed of (Cr3+ + V3+)
the sapphires crystallized in high-grade metamorphic and Fe3+-related features, as their iron concentration is
rocks (gneisses), or from magmas of highly alkaline com- moderate to high. GL
position, at shallower levels in the lithosphere than the
rocks that hosted the rubies. JES Host rock characteristics and source of chromium and
beryllium for emerald mineralization in the ophi-
Distinction of gem spinels from the Himalayan mountain olitic rocks of the Indus Suture Zone in Swat, NW
belt. A. Malsy ( and L. Pakistan. M. Arif, D. J. Henry, and C. J. Moon, Ore
Klemm, CHIMIA International Journal for Chemistry, Geology Reviews, Vol. 39, No. 1/2, 2011, pp. 1–20,
Vol. 64, No 10, 2010, pp. 741–746,
10.2533/chimia.2010.741. The Swat Valley region of northern Pakistan has been an
Gem spinel deposits in Myanmar (Mogok), Vietnam important source of emeralds for the past five decades. The
(Luc Yen), and Tajikistan formed in association with stones have been reported from a number of localities in
Himalayan orogenesis. Samples from these deposits this area, with the main deposits near Mingora and Gujar
were investigated by standard gemological testing and Kili. The emeralds are hosted by magnesite-talc-quartz-
LA-ICP-MS. In most of the Mogok spinels, microscopic dolomite assemblages that form part of the extensive
examination identified apatite and calcite inclusions sequence of metamorphosed sedimentary and igneous rocks
together with octahedral negative crystals. Dislocation along the tectonic collision zone between the Asian and
systems and sphene inclusions distinguished samples Indian continental plates. These magnesite-rich rocks are
from Luc Yen. A yellow surface fluorescence to short- mostly distributed along contacts between serpentinized
wave UV radiation was observed only in samples from ultramafic rocks and carbonate±graphite-bearing metasedi-
Tajikistan. Due to their similar formation conditions, mentary rocks. From their field association and petrographic
spinels from these sources display few origin-specific evidence, they apparently resulted from alteration by car-
characteristics in their trace-element pattern, but Ti, bonate-bearing metamorphic fluids. Late-stage hydrother-
Fe, Ni, Zn, Zr, and Sn differ slightly. The highest Zn mal activity in the fissile magnesite-rich rocks produced
concentrations were observed in Mogok material, quartz veins that locally contain emerald, Cr-tourmaline,
while samples from Luc Yen had the highest Fe con- and Cr,Ni-rich muscovite. Petrographic and mineralogical
tent. Spinels from these deposits can therefore be dis- analysis suggests that the Cr, Ni, and Mg in these minerals
tinguished by trace-element chemistry. This is espe- were derived by alteration of the original ultramafic rocks.
cially helpful for gems that show few or no inclusion In contrast, the Be and B appear to have been introduced by
features. GL hydrothermal solutions of igneous origin that invaded the
magnesite-rich host rocks. JES
Emeralds from South America—Brazil and Colombia. D.
Schwarz (, J. C. Laos: Land of a million elephants . . . and sapphires. R.
Mendes, L. Klemm, and P. H. S. Lopes, InColor, No. Hughes, A. Ishmale, F. Isatelle, and P. Wang,
16, Spring 2011, pp. 36–46. InColor, No. 16, Spring 2011, pp. 12–19.


Near the infamous Golden Triangle, the point where Laos, topaz, garnet (spessartine), and spodumene (kunzite).
Myanmar, and Thailand meet, lies the Ban Huay Xai sap- Although much has been written about the district and its
phire deposit. The deposit has been a point of geologic legendary discoveries, this article reports on recent finds
interest since 1890, and it saw a brief surge of activity in rivaling those of bygone decades. In particular, considerable
the 1960s as various mining companies took a passing gem rough and mineral specimens have emerged from the
interest in recovering its small rough material. Cyro-Genie and Oceanview mines in the past 10 years. The
In 2006, though, the Lao government partnered with region clearly is not depleted, though production is typical-
Sino Resources Mining Co., which staked a handful of ly cyclical with long intervals of inactivity. KAM
claims and is now obtaining 200 kg of rough sapphires per
month. Mining is done on a shallow level, with most Oscillatory zoned liddicoatite from Anjanabonoina, cen-
stones uncovered beneath 30 cm of topsoil and 1–1.5 m of tral Madagascar. II. Compositional variation and
overburden. The majority of the stones are small (<3 mechanisms of substitution. A. J. Lussier and F. C.
mm), but their color is exceptional. Hawthorne, Canadian Mineralogist, Vol. 49, No. 1,
Sino Resources is aware that its open-pit mining can 2011, pp. 89–104,
be destructive for the environment and the primarily canmin.49.1.89.
farming-based community. The company has adopted a For more than a century, granitic pegmatites at Anjana-
stance of stewardship and is restoring mined land to its bonoina (~60 km west of Antsirabe) have produced excep-
natural state, negotiating with farmers, and investing in tional color-zoned crystals of liddicoatite tourmaline.
training and employing the local community for more Slices of these crystals cut perpendicular to the c-axis that
than 95% of its labor force. AB exhibit banded, triangular-shaped, or other color zoning
patterns are much sought-after by both mineral collectors
Microstructure and origin of colour of chrysoprase from and jewelry manufacturers. This study was undertaken to
Haneti (Tanzania). H. A. Graetsch [heribert.graetsch@ analyze by electron microprobe one of these tourmaline], Neues Jahrbuch für Mineralogie, Abhandlun- slices from core to rim to better understand the changes in
gen, 2011, Vol. 182, No. 2, pp. 111–117, http://dx.doi. chemical composition that accompany the color zoning.
org/10.1127/0077-7757/2011/0187. Over the bulk of the crystal, the compositional varia-
The green color of chrysoprase is known to originate from tion corresponds to two substitution mechanisms:
a Ni-containing compound incorporated within the micro-
1. XNa + YM* Æ XCa + YLi
crystalline structure of chalcedony. However, the nature of
2. YM* + YM* Æ YLi + YAl
the admixed Ni-containing phase is not known precisely,
due to the lack of direct evidence from X-ray diffraction or where M* = Fe + Mg + Mn, and X and Y represent different
microscopy. crystallographic sites in the tourmaline structure.
To address this issue, the authors studied chrysoprase The analyses reveal a smooth monotonic variation in the
from the Haneti-Itiso area in central Tanzania, and com- concentrations of the principal constituents of the tourma-
parison samples from Poland and Australia, using optical line, and a superimposed oscillatory variation in several ele-
reflectance spectroscopy, near-infrared absorption ments—particularly Fe, Mg, Mn, Na, and Ca. This oscilla-
spectroscopy, and powder X-ray diffraction. The results tion is manifested as a decrease in the amount of an element
showed that the green color of Haneti chrysoprase is until a color zone boundary is reached, at which the concen-
caused by the incorporation of trace amounts of gaspeite (a tration rises abruptly, with this same pattern repeated at
Ni-carbonate mineral) in the microstructure of the chal- each boundary. The compositional and color variations are
cedony. Chrysoprase samples from the comparison different in the pyramidal and prismatic growth sectors. The
sources did not show any gaspeite diffraction peaks; their authors suggest specific chemical substitutions based on the
color seems to originate from a different pigment (i.e., two mechanisms shown above to explain the observed varia-
inclusions of poor crystallinity). The Haneti chrysoprase tions in chemical composition. They conclude that current
has a higher crystallinity of both the microcrystalline models for oscillatory zoning in minerals do not fully explain
quartz matrix and the gaspeite pigment. ERB the internal color patterns seen in this tourmaline, and a
more appropriate model needs to be developed. JES
Mines and minerals of the Southern California pegmatite
province. J. Fisher, Rocks & Minerals, Vol. 86, No. Raman microspectroscopy of organic inclusions in spo-
1, 2011, pp. 14–34, dumenes from Nilaw (Nuristan, Afghanistan). A.
00357529.2011.537167. Weselucha-Birczynska (,
Granitic pegmatites in Southern California have long yield- M. Słowakiewicz, L. Natkaniec-Nowak, and L.
ed a variety of gems, including some notable recent discov- M. Proniewicz, Spectrochimica Acta Part A: Molec-
eries. This nontechnical summary provides an excellent ular and Biomolecular Spectroscopy, Vol. 79, No. 4,
mining history of the region. It is liberally illustrated with 2011, pp. 789–796,
impressive specimens of tourmaline, beryl (morganite), j.saa.2010.08.054.


Spodumene crystals from the Nuristan region are among JEWELRY HISTORY
the finest examples of this mineral ever found. Gem spo-
Emeralds, sapphires, pearls and other gemmological mate-
dumene occurs in two main color varieties in Afghanistan:
rials from the Preslav gold treasure (X century) in
pink to violet (kunzite) and yellowish green. Samples from
Bulgaria. E. Strack ( and R. I.
the Nilaw mine were investigated by microthermometry
Kostov, Bulgarian Academy of Sciences, Bulgarian
and Raman spectroscopy. Microthermometry of the pri-
Mineralogical Society, Vol. 48, 2010, pp. 103–123.
mary fluid inclusions indicated crystallization ranges of
The Preslav gold treasure was found in 1978 near the town
300–650°C and 1.5–2.5 kbar. The brine concentration in
of Veliki Preslav, the second Bulgarian capital during the
the fluid inclusions varied from 4.3 to 6.6 wt.% NaCl
end of the First Bulgarian Kingdom. The 10th-century trea-
equivalent salinity. Raman spectra of selected fluid, organ-
sure belonged to a female member of the royal family and
ic, and solid inclusions were collected in linear or rectan-
is believed to be of Byzantine origin. Among the gem mate-
gular arrays as well as depth profiles to study their size and
rials identified are emerald, violet sapphire, pearl, reddish
contents. The phases documented were calcite, beryl,
violet garnet, rock crystal, amethyst, and carnelian. The
topaz, and spodumene accompanied by fluid and/or organ-
average dimensions of the emeralds in two medallions
ic inclusions (liquid and gas hydrocarbons) with bands at
were 0.5 cm long and 0.6 cm wide; their origin is suspected
2900 cm-1 (C2H6-CH3), 2550 cm-1 (H2S), and 2350 cm-1
to be Egyptian or Austrian based on the deposits known at
(CO2, N2). Also confirmed was the presence of carbona-
the time and similarities in their morphology, internal fea-
ceous matter (D-band at ~1320 cm-1 and/or G-band at
tures, and inclusions. Fine rutile silk inclusions were iden-
~1600 cm-1), which developed in the course of transforma-
tified in the sapphire. As the Indian subcontinent was the
tion of the original hydrocarbons. GL
main source of gem corundum in antiquity and Early
Medieval times, it can be assumed that the sapphires in the
Preslav treasure are of an Indian or Sri Lankan origin. The
INSTRUMENTS AND TECHNIQUES barrel- and baroque-shaped saltwater pearls (average 0.4 cm
Use of the Raman spectrometer in gemological laborato- long and 0.5 cm in diameter), some of them partly decom-
ries: Review. L. Kiefert and S. Karampelas (s.karam- posed, are thought to originate from the Persian Gulf and, Spectrochimica Acta Part the coastal areas of the Indian Ocean. GL
A: Molecular and Biomolecular Spectroscopy, Vol.
80, No. 1, 2011, pp. 119–124,
10.1016/j.saa.2011.03.004. JEWELRY RETAILING
The Raman effect was first reported in 1928, and for Watches and jewellery retailing, executive summary – UK –
decades it was used mainly in physics and chemistry. September 2011. Mintel,, 10 pp.
With the availability of Raman microscopes in the mid- This report, available only to Mintel subscribers, forecasts
1970s, the technique became better suited for geosciences, sales of watches and jewelry (not including costume jewel-
including gemology. In the 1990s, Raman spectrometers ry) in the UK at £4.2 billion (US$6.6 billion) in 2011, a
became smaller, more sensitive, and more affordable, growth of 3% since 2006. The value of the market shrank
expanding their use in gemological laboratories. The tech- by 1% between 2007 and 2009. Mintel suggests that sales
nique is now routinely used to establish the identity of an will increase by 8% between 2011 and 2016 to reach £4.6
unknown gem, study inclusions, identify treatments, and billion (US$7.2 billion), but there are challenges for both
characterize natural or artificial color in pearls and corals. the watch and jewelry sectors.
Raman spectroscopy offers a nondestructive means of The watch market faces competition from portable
gem identification that does not require any contact with electronic devices that display the time. Based on their
the sample. Identification of subsurface inclusions is also consumer research, Mintel calculates the average price
possible in some cases. In the area of gem treatments, the people would spend on a watch for themselves in 2011 is
method is used for the detection of emerald fillers; the iden- £87, a notable 17% decline from £102 in 2010.
tification of lead-, barium-, or zirconium-rich glass-fillings The precious metal jewelry market, which accounts
(or the presence of fluxes) in corundum; and the detection of for three-quarters of market value, is affected by the soar-
heat-treated corundum and HPHT-annealed diamonds. ing gold price, which has decreased volume sales. The
The authors conclude that Raman spectroscopy has volatile price of gold is the main reason behind a 77%
become an indispensable tool for gemological laboratories, reduction in the weight of gold pieces hallmarked in UK
and that many new applications are waiting to be found. assay offices between 2001 and 2010.
Future possibilities include the use of the Raman photolu- Marriages have ebbed to the lowest rate since records
minescence spectroscopy for characterizing colored dia- began nearly 150 years ago, according to the UK’s Office
monds and studying the role of rare-earth elements in col- of National Statistics. Demand for engagement and wed-
ored stones. DAZ ding rings have therefore dropped. But gift giving and


receiving remains popular, and the permanence of watch- scale, bringing little direct benefit to local government,
es and jewelry will continue to drive sales. EJ while environmental damage and the effects of overpopula-
tion during an emerald “rush” can create lasting problems.
And without tax revenues from mining, local governments
have no incentive to offer improved health and living stan-
PRECIOUS METALS dards to miners. The authors propose several solutions to
Determination of gold in jewellery alloys via potentiomet- mitigate these conditions, including: (1) bringing in outside
ric titration: Procedure standardization and results investors who are accountable to international standards,
benchmarking. G. Pezzatini, M. Caneschi, M. Inno- and possess technology for more efficient and less destruc-
centi, S. Bellandi, E. Lastraioli, L. Romualdi, and S. tive mining; (2) introducing corporate governance stan-
Caporali, Current Analytical Chemistry, Vol. 7, No. dards; (3) encouraging miners’ cooperatives that will pay
4, 2011, pp. 277–285, taxes and provide incentive for governments to create ser-
157341111797183065. vices; and (4) developing local markets to realize higher
Potentiometric titration is a volumetric method in which returns. RS
the electrical potential between two solutions is measured
as a function of the added reagent volume. The authors Potential for modernization of small-scale gemstone min-
recommend this method for the accurate determination of ers—The case of the tsavorite mines in southeast
the gold content of jewelry alloys rather than traditional Kenya. C. Simonet, InColor, No. 16, Spring 2011,
cupellation (i.e., fire assay), which presents serious health pp. 22–27.
and environmental risks.
A wide variety of gold jewelry alloys were dissolved in The exact location of colored stone mineralization is
acid, and the resulting solutions were used in an optimized hard to predict, even by applying drilling methods. Thus
titration procedure developed by the authors. To evaluate it is extremely difficult to mechanize the mining of col-
the proposed method, the authors performed experiments in ored stones, especially for primary deposits. The low
three different laboratories. The results were consistent percentage of high-grade material produced from most
with data obtained by cupellation, demonstrating that deposits also contributes to the inability to sustain
potentiometric titration is a robust and affordable alterna- mechanized mining. Thus, gem miners are still using
tive to classic fire assay for testing gold alloy jewelry. crude methods that do not meet international standards.
JES Kenya is one of the leading producers of tsavorite.
Mining is mostly done with basic equipment, such as
compressors and jackhammers. Challenges include the
Microstructure analysis of selected platinum alloys. P. Batta- lack of technical knowledge, the irregularly spaced tsa-
ini, Platinum Metals Review, Vol. 55, No. 2, 2011, pp. vorite concentrations, and the steeply dipping mineral-
74–83, ized zones, resulting in a series of narrow tunnels. As a
Optical metallography involves the microscopic examina- consequence, gem production is unpredictable and the
tion of polished or etched surfaces of metal. The technique prices are low. Kenya has many tsavorite deposits with
can be employed to describe a metal’s microstructures both similar structural geometry and geologic characteristics,
qualitatively and quantitatively. It is often supplemented and this presents a good opportunity to modernize
by other methods, such as SEM-EDS analysis, to more fully small-scale mining. This initiative would require stan-
characterize a material. In this article, the author describes dard mining procedures, properly defined equipment,
the metallographic analysis of the internal microstructure and detailed geologic surveys using remote sensing, geo-
of platinum metal alloys similar to those used for jewelry physics, and geochemistry. Other aspects include train-
and industrial applications. A number of examples illus- ing miners; providing expertise from geologists, gemolo-
trate the visual distinction between these various alloys, gists, and mine engineers; improving access to funding;
and whether the alloy sample was merely crystallized (“as and branding. Modernization of tsavorite mining would
cast”) or had been work-hardened and annealed. JES promote market stability and compliance with proper
health, safety, and environmental standards.
In recent years, two major gem mining companies
MISCELLANEOUS have been listed on stock exchanges. This ensured their
compliance with international standards of scientific
Gemstone mining as a development cluster: A study of work, mining methods, planning, and governance,
Brazil’s emerald mines. J. A. Puppin de Oliveira, S. including publication of resource estimates and techni-
Ali, Resources Policy, Vol. 36, 2011, pp. 132–141. cal studies. This trend is an encouraging sign for the
Most Brazilian emerald mining is clustered in three areas modernization of the colored stone mining industry.
within the country. The vast majority of mining is small KSM


Volume 47

This index gives the first author (in parentheses), issue, and inclusive pages of the article in which the subject occurs for all feature
articles, Notes & New Techniques, and Rapid Communications that appeared in Volume 47 of Gems & Gemology. Also included
are abstracts of presentations from the Summer 2011 (S11) issue, the proceedings of the Fifth International Gemological
Symposium. For the Gem News International (GNI) and Lab Notes (LN) sections, inclusive pages are given for the item. The Book
Reviews section is available only as an online supplement, with the page numbers preceded by S. The Author Index (p. S19) pro-
vides the full title and coauthors of the articles cited.

A Aquamarine simulants
synthetic quartz (GNI)W11:332-333
Book reviews
Collectors Guide to Granite
Afghanistan Argon isotope analysis Pegmatites (King)F11:S4
afghanite from Badakhshan of andesine, from Tibet and other Collectors Guide to Silicate Crystal
(GNI)F11:235 claimed localities (Rossman)Sp11:16- Structures (Lauf)F11:S4
emerald mining in (GNI)F11:238-239 300 Colour of Paradise: The Emerald in the
sapphire from, beryllium- and tungsten- Age of Gunpowder Empires
Argyle diamond mine
bearing (LN)Sp11:53-54 (Lane)Sp11:S2-S3
colored diamonds from
scapolite from (GNI)Sp11:65-66 (Chapman)Su11:130 The Diamond Compendium
sphene from border area with Pakistan, (Cunningham)W11:S1
color-change (GNI)W11:327-328 Assembled gem materials Diamond Math (Glasser)F11:S4
agate and reconstituted turquoise dou- Diamonds in Nature: A Guide to
Afghanite blet, marketed as Coral Sea agate
from Afghanistan (GNI)F11:235 Rough Diamonds (Tappert and
(GNI)Sp11:62 Tappert)W11:S2
Agate corundum and lead-glass triplet The Extraordinary World of Diamonds
dendritic, in jewelry (GNI)Sp11:62-63 (GNI)F11:251-252 (Norman)F11:S1-S2
and reconstituted turquoise doublet, glass triplet imitation of “mystic” Gems and Gemology in Pakistan
marketed as Coral Sea (GNI)Sp11:62 treated topaz (GNI)F11:252-253 (Khan and Kausar)F11:S4
Amethyst opal with artificial matrix Jewellery from the Orient: Treasures
separation from synthetic, using (LN)W11:312 from the Bir Collection
infrared spectroscopy Asterism (Siewert)Sp11:S3
(Karampelas)F11:196-201 artificial (Steinbach)Su11:152-153 Living Jewels: Masterpieces from
from southern Brazil in quartz, caused by cinnabar inclu- Nature (Peltason)F11:S1
(Juchem)Su11:137-138 sions (GNI)W11:322-323 Mineral Treasures of the World (The
Amethyst, synthetic Auctions Geological Museum of China and
separation from natural, using infrared of luxury jewelry (Luke)Su11:100-102 The Collector’s Edge Minerals
spectroscopy (Karampelas)F11:196- Australia Inc.)F11:S4
201 colored diamonds from Argyle Pearls (Bari and Lam)Sp11:S1
Ametrine [amethyst-citrine] (Chapman)Su11:130 Precious Objects: A Story of
from Bolivia, Anahí mine opal from—Laverton, Western Diamonds, Family, and a Way of
(Weldon)Su11:163-164 Australia (GNI)W11:319-320; Life (Oltuski)W11:S1-S2
Ammonite nomenclature and characterization Terra Spinel Terra Firma (Yavorsky and
iridescent, from Madagascar (Beattie)Su11:116 Hughes)Sp11:S1-S2
(GNI)F11:235-236 zircon from, exhibiting tenebrescence Twentieth-Century Jewellery: From Art
(LN)W11:314-315 Nouveau to Contemporary Design in
Analytical techniques [general] Europe and the United States
to characterize gem materials (Cappellieri)Sp11:S3
What’s Hot in Tucson 2010 (Blue Cap
Andesine B Productions)W11:S2
reportedly from Tibet (Rossman)Sp11:16- The Workbench Guide to Jewelry
30, (Abduriyim)Su11:167-180 Beryl, see Aquamarine, Emerald
Techniques (Young)F11:S2
Andradite Beryllium World Hallmarks, Volume I: Europe,
demantoid from northern Madagascar natural, in blue sapphire—(LN)F11:232- 19th to 21st Centuries, 2nd ed.
(Pezzotta)Sp11:2-14 233; from Afghanistan (LN)Sp11:53- (Whetstone, Niklewicz, and
54 Matula)F11:S3-S4
Annealing, see Diamond treatment
Aquamarine Beryllium diffusion, see Diffusion treatment Boxes [article sidebars]
from Thanh Hoa, Vietnam Bolivia andesine from Tibet and other claimed
(Huong)Sp11:42-48, (GNI)F11:236- ametrine from the Anahí mine localities—feldspar nomenclature
237 (Weldon)Su11:163-164 (Rossman)Sp11:16-30; samples and


experimental details development and tax policy of gem (Choudhary)Su11:145-146
(Rossman)Sp11:16-30 industry in (Jia)Su11:159-160 market for (Drucker)Su11:158-159
diamond symmetry—basic measuring freshwater cultured pearls from—bead- marketing of (Overlin)Su11:97-99
concepts (Geurts)W11:286-295; com- ed with baroque freshwater cultured from Mexico, review of
binations of symmetry features pearls (GNI)F11:244-245; with exotic (Ostrooumov)Su11:141
(Geurts)W11:286-295 metallic colors (Beavers)Su11:144; world production of (Yager)Su11:142-143
magnetic susceptibility—and magnetic treatments and their detection
Conch pearl
materials (Hoover)W11:272-285; mea- (Strack)Su11:120
large (LN)F11:230-231
surement of (Hoover)W11:272-285 nephrite from, natural and treated
(Zhang)Su11:122 Conferences
Brazil ICA Congress 2011 (GNI)F11:253-254
diamond from, black, nature of col- sapphire, blue and pink, from Muling
(Chen)Su11:136-137 International Gemmological
oration (Vasilyev)Su11:135
topaz, neutron-irradiated blue, from Conference (GNI)W11:334-335
gemstones from southern
Guangdong (Zhang)W11:302-307 International Gemological Symposium
zircon, brownish red, from Muling (Proceedings issue)Su11:79-166
opal (or cristobalite?) from, blue
(Chen)Sp11:36-41 NAG Institute of Registered Valuers
Conference (GNI)W11:335
spodumene from, green cat’s-eye Chondrodite
Scottish Gemmological Conference
(GNI)F11:249-250 from Tanzania (GNI)W11:316
2011 (GNI)F11:254
tourmaline from Pederneira, Minas Chrysocolla, see Chalcedony Sinkankas Symposium 2011—
Gerais (Pezzotta)Su11:141-142
Citrine simulants Diamond (GNI)F11:254-255
Burma, see Myanmar large synthetic quartz (GNI)W11:332- Conflict diamonds
333 debate on (Symposium Debate
Clinohumite Center)Su11:126-128
C unusually large (LN)F11:222-223 Congo, Democratic Republic of the
California, see United States Coating andesine reportedly from
on bead-cultured freshwater pearls (Rossman)Sp11:16-30
Canada (LN)W11:313-314 Conoscopy
diamonds from (Stachel)Su11:112-114 on diamond—black (LN)F11:223-224; technique for identifying interference
Carving, see Lapidary arts with spectroscopic features of a natu- figures in gem materials
Cat’s-eye, see Chatoyancy ral-color pink (LN)F11:224-225 (Gumpesberger)Su11:147
“Cat’s-eye pearls” Colombia Coral
non-nacreous calcitic pearl imitations dolomite, blue, from Muzo conservation of (Carmona)Su11:158
(GNI)W11:330-332 (GNI)W11:316-318
Coral Sea agate
Cavansite Color, cause of agate and reconstituted turquoise dou-
faceted (GNI)Sp11:56-57 in natural-color and treated colorless blet (GNI)Sp11:62
Chalcedony and colored diamonds
purplish blue and red-brown, from Peru (Epelboym)Su11:133
mining in Pakistan (GNI)W11:325-326
(GNI)F11:237-238 in tanzanite (Smith)Su11:119-120
triplet with lead glass (GNI)F11:251-
in tourmaline from Mt. Marie, Maine
“Challenge,” see Gems & Gemology 252
see also Ruby, Sapphire
Chatoyancy Color change
in quartz—caused by tourmaline needle Country of origin, see Geographic origin
sphene, vanadium-bearing, from
inclusions (GNI)F11:245-246; with Pakistan/Afghanistan (GNI)W11:327- Cristobalite
cinnabar inclusions (GNI)W11:322- 328 or blue opal(?) from Brazil
323 (Schnellrath)Su11:142
Color grading
in spodumene, green (GNI)F11:249-250 Cultured pearl, see Pearl, cultured
of diamond, nomenclature
Chemical composition (Tashey)Su11:163 CVD [chemical vapor deposition]-grown
of andesine, reportedly from Tibet dual integrating sphere spectrometer synthetic diamonds, see Diamond,
(Rossman)Sp11:16-30, with artificial intelligence synthetic
(Abduriyim)Su11:167-180 (Liu)Su11:154
of demantoid from northern
Color zoning
Madagascar (Pezzotta)Sp11:2-14
of garnet—(Henderson)Su11:148; deter-
in andesine, reportedly from Tibet D
mined using magnetic susceptibility (Abduriyim)Su11:167-180 Demantoid, see Andradite
(Hoover)W11:272-285 Dentine, see Odontolite
in low-grade diamonds (Kwon)Su11:134
of zircon, brownish red, from China
in opal, bicolored (GNI)W11:320-322 Diamond
in topaz, yellowish to greenish brown brooch from September 11 attacks
Chemical vapor deposition [CVD], see (GNI)F11:250-251 (GNI)F11:255-256
Diamond, synthetic cleaning technology (Vins)Su11:136
Colored stones
Chemometrics from Brazil, southern color grading of, nomenclature
of corundum, in determining geograph- (Juchem)Su11:137-138 (Tashey)Su11:163
ic origin (Yetter)Su11:157 discriminant analysis for identifying color origin, determination of
China geographic origin of (Epelboym)Su11:133
andesine, reportedly from Tibet (Blodgett)Su11:145 color zoning in low-grade diamonds
(Rossman)Sp11:16-30, examined at the Gem Testing (Kwon)Su11:134
(Abduriyim)Su11:167-180 Laboratory in Jaipur, India crystal descriptions, for use in explo-


ration (Klettke)Su11:139 (Willems)F11:202-207; optical proper-
ties of silicon-related defects
inclusions in low-grade diamonds
(Kwon)Su11:134 (D’Haenens-Johansson)Su11:131-132 Economy
length-to-width ratios among fancy- HPHT-grown—donor nitrogen aggrega- global, GIA Symposium keynote
shapes (Blodgett)Su11:129 tion in (Vins)Su11:135-136; pink address on (Overlin)Su11:80-82
localities, alluvial (Janse)Su11:110 (Johnson)Su11:133-134 Editorials
with luminescent cleavage marketing of (Overlin)Su11:97-99 “GIA Symposium 2011: Advancing the
(LN)F11:226-227 treated, with pink color intensified by Science and Business of Gems”
marketing of (Overlin)Su11:97-99 fluorescence (LN)F11:228-229 (Kimmel and Keller)Su11:79
mining of—(Coopersmith)Su11:108; in Diamond treatment “Great Expectations” (Iverson)W11:259
Canada (Stachel)Su11:112-114; in characterization of (Epelboym)Su11:133 “New Beginnings” (Iverson)F11:181
Liberia (GNI)Sp11:63-64 coating—black (LN)F11:223-224; with EDXRF, see Spectroscopy, energy-disper-
phosphorescence, long-lasting spectroscopic features of a natural- sive X-ray fluorescence
(LN)W11:310-311 color pink (LN)F11:224-225
polished supplies from secondary mar- Electron-microprobe analysis
HPHT—discriminant analysis in dis-
ket sources (Shor)Su11:163 of andesine from various claimed local-
tinguishing (Blodgett)Su11:145; with
ities (Rossman)Sp11:16-30
postage stamp collection fluorescence pattern of an HPHT-
(Overlin)F11:214-219 of aquamarine from Vietnam
grown synthetic (LN)W11:310; large
from Russia’s Nurbinskaya pipe, min- brownish yellowish orange
eralogical characteristics of (LN)Sp11:49; large colorless of demantoid from northern
(Solodova)Su11:134-135 (LN)Sp11:49-50; large pink Madagascar (Pezzotta)Sp11:2-14
strain in colorless untreated (LN)F11:227; observation of strain in of ruby and sapphire from Vietnam
(LN)F11:224-225 detecting (Eaton-Magaña)Su11:132; (Khoi)F11:182-195
symmetry grading of—assessments and subtle color enhancement of zircon, brownish red, from China
metrics (Caspi)Su11:153-154; GIA (LN)F11:225-226 (Chen)Sp11:36-41
boundaries (Geurts)W11:286-295 irradiation, producing green and orangy see also Chemical composition
thermoluminescence in type IaB brown spots (Nasdala)Su11:105-106 Emerald
(LN)Sp11:50 of synthetics—donor nitrogen aggrega- from Afghanistan, mining update
with 3H and H3 defects resulting from tion in (Vins)Su11:135-136; pink (GNI)F11:238-239
radiation damage (Choi)Su11:131 (Johnson)Su11:133-134; (LN)F11:228- formation of (Giuliani)Su11:108-110
from Zimbabwe, naturally irradiated 229 geographic origin of (de Corte)Su11:147
(Breeding)Su11:129-130, DiamondView imaging inclusions in quartz (GNI)W11:323
(Crepin)Su11:105 of CVD-grown synthetic diamonds, market for (Drucker)Su11:158-159
see also Diamond, colored; Diamond, luminescent regions in separation from synthetic (de
cuts and cutting of; Diamond, inclu- (Willems)F11:202-207 Corte)Su11:147
sions in; Diamond, synthetic; of HPHT-grown pink synthetic dia- from Zambia, mining of (GNI)Sp11:63-65
Diamond treatment; DiamondView monds (Johnson)Su11:133-134 Emerald simulants
imaging of HPHT-treated diamond with fluores- plastic-coated quartz (GNI)Sp11:71-72
Diamond, colored cence pattern of an HPHT-grown trapiche (GNI)Sp11:68-69
from the Argyle mine, studies of synthetic (LN)W11:310
Emerald, synthetic
(Chapman)Su11:130 Diffusion treatment separation from natural (de
black—from Brazil, nature of coloration experiments performed on andesine Corte)Su11:147
(Vasilyev)Su11:135; coated (Rossman)Sp11:16-30,
(LN)F11:223-224; colored by strong Enhancement, see Coating; Diamond treat-
plastic deformation (LN)F11:223; ment; Diffusion treatment; Dyeing;
of sapphire with Be—durability and Filling, fracture or cavity; Heat treat-
strongly purple-colored (LN)W11:309 safety testing (Sutthirat)Su11:121; ment; Irradiation; Treatment; specific
color grading of, nomenclature experiments (Wathanakul)Su11:153; gem materials
(Tashey)Su11:163 pink (LN)F11:232
color origin, determination of EPR, see Spectroscopy, electron paramag-
Discriminant analysis netic resonance
(Epelboym)Su11:133 of andesine, reportedly from Tibet
Fancy Vivid purple (LN)W11:308 (Abduriyim)Su11:167-180 Errata
greenish brown, with a color shift to “CVD synthetic diamond lumines-
to identify geographic origin of colored
(LN)F11:234-235 cent regions” (Willems)F11:202-207—
stones and HPHT treatment in dia-
incorrect photographer (GNI)W11:335
see also Diamond, synthetic; Diamond monds (Blodgett)Su11:145
treatment to “Extreme conoscopy”
Dolomite (Gumpesberger)Su11:147—incorrect
Diamond, cuts and cutting of blue, from Colombia (GNI)W11:316-318 caption (GNI)F11:256
to exhibit inclusions (LN)Sp11:50-52 Doublets, see Assembled gem materials to “Gem treatments retrospective”
preferences in length-to-width ratios Durability (McClure)F10:218-240—description
(Blodgett)Su11:129 of Be-diffused sapphire of topaz treatment methods
Diamond, inclusions in (Sutthirat)Su11:121 (GNI)Sp11:73
in low-grade diamonds (Kwon)Su11:134 of lead-glass-filled ruby to “Natural-color tanzanite”
manufactured to exhibit (GNI)Sp11:50-52 (Sutthirat)Su11:121 (Smith)Su11:119—incorrect caption
3D mapping systems (Caspi)Su11:153 Dyeing (GNI)F11:256
Diamond, synthetic of Chinese freshwater cultured pearls, to “Sapphire and zircon from Ethiopia”
chemical vapor deposition [CVD]— detection of (Strack)Su11:120 (GNI)F11:247-248—errors in absorp-
from Gemesis (LN)F11:227-228; of opal from Ethiopia, purple tion peak assignments (GNI)W11:335
luminescent regions in (Renfro)W11:260-270 to “Vietnamese ruby and sapphire”


(Khoi)F11:182-195—incorrect caption Gems & Gemology gemstones (Eickhorst)Su11:123
(GNI)W11:335 “Challenge”—Sp11:75-76; winners and Imitations, see specific gem materials
Ethics answers F11:220 imitated
debate on (Symposium Debate Edward J. Gübelin Most Valuable Inclusions
Center)Su11:126-128 Article Award Sp11:1-2 in andesine, reportedly from Tibet
Ethiopia Geographic origin (Rossman)Sp11:16-30,
opal from Wollo—black (LN)W11:312- of colored gemstones, determination of (Abduriyim)Su11:167-180
313; dyed purple (Renfro)W11:260- (Abduriyim)Su11:114-116 of cinnabar in chatoyant quartz
270; play-of-color of corundum, determination using LIBS (GNI)W11:322-323
(Rondeau)Su11:112; sugar-acid treat- and advanced chemometrics in dyed purple opal from Ethiopia
ment of (GNI)W11:333-334 (Yetter)Su11:157 (Renfro)W11:260-270
sapphire from (GNI)F11:247-248 discriminant analysis, application of in of emerald in quartz (GNI)W11:323
zircon from (GNI)F11:247-248 determining—(Blodgett)Su11:145; of and geographic origin of colored gem-
andesine (Abduriyim)Su11:167-180; stones (Abduriyim)Su11:114-116
of emerald (de Corte)Su11:147 of izoklakeite in quartz skull carving
F Materialytics Sequencing System (GNI)W11:324-325
(McManus)Su11:155-156 of spessartine in quartz (GNI)F11:245
Faceting, see Diamond, cuts and cutting of of peridot (Thoresen)Su11:121-122 of tourmaline in cat’s-eye quartz
Fair trade practices Geology (GNI)F11:245-246
debate on (Symposium Debate of Canadian diamond deposits of trolleite and lazulite in blue quartz
Center)Su11:126-128 (Stachel)Su11:112-114 (GNI)Sp11:57-58
examples and case studies see also Diamond, inclusions in
(Luke)Su11:103-104 Glass
triplet imitation of “mystic” treated India
Fakes, see specific gem materials simulated topaz (GNI)F11:252-253 agate from, dendritic (GNI)Sp11:62-63
Feldspar Grossular Gem Testing Laboratory in Jaipur,
sunstone from Oregon, comparison of green, from Tanzania (GNI)Sp11:57-58 remarkable gems encountered at
two deposits (McClure)Su11:150-151 (Choudhary)Su11:145-146
see also Andesine Guatemala
quartz from, with spectral interference
jadeite from, lavender (Harlow)Su11:116
Filling, fracture or cavity (GNI)Sp11:58-59
of ruby with a lead glass—durability Gübelin gem collection scapolite from (GNI)Sp11:59
and safety testing (Sutthirat) online resource of (Shigley)Su11:151-152
Infrared spectroscopy, see Spectroscopy,
Su11:121; trapiche (GNI)Sp11:72 infrared
Fluorescence, ultraviolet [UV] H Instruments
applications in gemology dual integrating sphere for color grading
(Taylor)Su11:125 Halite and measurement (Liu)Su11:154
of an HPHT-treated diamond, with pat- submitted for identification inclusion mapping system for dia-
tern resembling an HPHT-grown (LN)Sp11:52-53 monds (Caspi)Su11:153
synthetic diamond (LN)W11:310 Hallmarking spectral analysis, automated real-time
Fluorite of precious metals, benefits of (Magaña)Su11:154-155
from Namibia (GNI)W11:318-319 (Niklewicz)Su11:160-161 symmetry grading system for diamonds
France Heat treatment (Caspi)Su11:153-154
odontolite (fossilized mastodon ivory) of Chinese freshwater cultured pearls, ultra-deep diamond cleaning technolo-
from (Krzemnicki)W11:296-301 detection of (Strack)Su11:120 gy (Vins)Su11:136
of odontolite, blue, in antique brooches see also Argon isotope analysis;
Fuchsite Chemometrics; Conoscopy;
deep green rock (GNI)F11:239-240 (Krzemnicki)W11:296-301
of ruby, pretreatment (Lee)Su11:148-149 DiamondView imaging; Electron-
of zircon, brownish red, from China microprobe analysis; Fluorescence,
ultraviolet; Illumination methods;
G (Chen)Sp11:36-41
Magnetic susceptibility; Microscopy;
High-pressure, high-temperature [HPHT] Oxygen isotope analysis; Scanning
synthesis, see Diamond, synthetic electron microscopy; Spectralysis;
chemical composition, determination
of—using Raman spectroscopy High-pressure, high-temperature [HPHT] Spectrometer; Spectrometry [vari-
(Henderson)Su11:148; using magnetic treatment, see Diamond treatment ous]; Spectroscopy [various]; X-radio-
susceptibility (Hoover)W11:272-285 History graphy; X-ray diffraction
color-change, from Tanzania of diamonds, in themed postage stamp Interference
(GNI)F11:240-241 collection (Overlin)F11:214-219 spectral, in quartz from India
hydrogrossular, green cabochon resem- Hydrogrossular, see Garnet (GNI)Sp11:58-59
bling jadeite (LN)Sp11:52-53 Hydrophane International Gemological Symposium,
see also Andradite, Grossular bicolor, bi-pattern opal (GNI)W11:320- 5th [2011]
Gemesis 322 proceedings of (Kimmel and
CVD-grown synthetic diamonds dyed purple opal from Wollo, Ethiopia Keller)Su11:79
(LN)F11:227-228 (Renfro)W11:260-270 Internet
Gemological Institute of America and jewelry retailing (Overlin)Su11:95-96
Gübelin gem collection, online Irradiation
resource (Shigley)Su11:151-152 I of diamond—producing green and
Symposium 2011 (Kimmel and Illumination methods orangy brown spots on surface
Keller)Su11:79 in the buying, selling, and grading of (Nasdala)Su11:105-106; producing 3H


and H3 defects (Choi)Su11:131 tourmaline from, correlation of man- Moissanite, synthetic
natural, of diamonds from Zimbabwe— ganese and iron (Ahn)Su11:143-144 with crystalline silicon intergrowth
(Crepin)Su11:105; hydrogen-rich Magnetic susceptibility (LN)Sp11:54-55
(Breeding)Su11:129-130 as a means of determining garnet com- Moonstone
of sapphire, with electron-beam position (Hoover)W11:272-285 synthetic star spinel imitation of
(Seo)Su11:151 Maine, see United States (LN)Sp11:54-55
of topaz, blue (Zhang)W11:302-307 Most valuable article, see Gems &
Marketing and distribution
Isotopes, see Argon isotope analysis, abbreviated jewelry descriptions Gemology
Oxygen isotope analysis (Root)Su11:161-162 Myanmar
Ivory of ametrine from the Anahí mine, colorless gems from (GNI)F11:241-242
mastodon, fossilized (odontolite), heat- Bolivia (Weldon)Su11:163-164 gem sales and production
ed blue (Krzemnicki)W11:296-301 and China’s gem industry (GNI)Sp11:72-73, (GNI)F11:256
Izoklakeite (Jia)Su11:159-160 jadeite from, lavender
inclusions in quartz skull carving of colored stones (Drucker)Su11:158- (Harlow)Su11:116
(GNI)W11:324-325 159, (Overlin)Su11:97-99 ruby overgrowth on painite, from
of coral jewelry (Carmona)Su11:158 Mogok (Nissinboim)Su11:140-141
of cultured pearls (Overlin)Su11:97-99 spinel, trapiche, from Mogok
J of diamonds—(Overlin)Su11:97-99; (GNI)W11:329-330
color grading (Tashey)Su11:163; from
Jade, see Jadeite, Nephrite secondary market sources
Jadeite (Shor)Su11:163 N
green cabochon (LN)Sp11:52-53 digital-age (Overlin)Su11:95-96
lavender, from Myanmar, Guatemala, hallmarking of precious metals, bene- Namibia
and Japan (Harlow)Su11:116 fits of (Niklewicz)Su11:160-161 fluorite from Klein Spitzkoppe region
sales at the 2010 Myanmar Gem lectures, gem and jewelry (Serras- (GNI)W11:318-319
Emporium (GNI)Sp11:72-73 Herman)Su11:162 Nassau, Kurt
Japan in the mineral collector market obituary (GNI)Sp11:73
jadeite from, lavender (Harlow)Su11:116 (Johnson)Su11:160 Nephrite
Jewelry of ruby—perception of rarity from China, natural and treated
antique and estate, selling of (Robertson)Su11:161; “pigeon’s (Zhang)Su11:122
(Luke)Su11:100-102 blood” (Atichat)Su11:157-158 rough, imitation of (GNI)Sp11:69-70
design overview (Luke)Su11:92-94 Materialytics Sequencing System Neutron irradiation, see Irradiation
luxury trends (Overlin)Su11:89-91 quantitative process to determine geo- Nomenclature and classification
graphic origin of gem materials abbreviated jewelry descriptions
(McManus)Su11:155-156 (Root)Su11:161-162
L Mauritania of diamond—color grades
Nuummite from (GNI)F11:242-243 (Tashey)Su11:163; crystals, for use
LA-ICP-MS, see Spectrometry, laser abla-
Measurement in geology and exploration
tion–inductively coupled plasma–mass
concepts and uncertainty of (Klettke)Su11:139
(Geurts)W11:286-295 “pigeon’s blood” ruby, standards for
Lapidary arts (Atichat)Su11:157-158
quartz skull carving with inclusions of Meteorites
pallasite with gem-quality peridot— “royal blue” sapphire, standards for
izoklakeite (GNI)W11:324-325 (Atichat)Su11:157-158
rhodochrosite carving of ancient Indian (Shen)F11:208-213; mounted in jew-
elry (GNI)F11:243-244 of ruby, effect on perception of rarity
sage (GNI)F11:246-247 (Robertson)Su11:161
see also Diamond, cuts and cutting of Mexico
colored gemstones from, review of Nuummite
Liberia from Mauritania (GNI)F11:242-243
diamond mining in (GNI)Sp11:63-64
opal, blue, from Sinaloa (GNI)F11:243
ruby from (Kiefert)Su11:138
LIBS, see Spectroscopy, laser-induced
polarized light and fluorescence, to O
establish origin of color in natural Obituary
Lighting diamonds (Epelboym)Su11:133
and the buying, selling, and grading of Kurt Nassau (GNI)Sp11:73
Mining and exploration
gemstones (Eickhorst)Su11:123 Odontolite
of colored gemstones worldwide
Luminescence fossilized dentine (mastodon ivory),
of gem materials (Fritsch)Su11:123, blue, identified in antique brooches
concrete mixer to sort gem-bearing (Krzemnicki)W11:296-301
(Rossman)Su11:124-125 gravels (Limsuwan)Su11:139
see also DiamondView imaging; of corundum in Pakistan Oiling
Fluorescence, ultraviolet [UV] (GNI)W11:325-326 of opal with artificial matrix
for diamond—(Coopersmith)Su11:108; (LN)W11:312
in Canada (Stachel)Su11:112-114; in Omphacite
M Liberia (GNI)Sp11:63-64 green cabochon resembling jadeite
Madagascar for emerald—(Giuliani)Su11:108-110; (LN)Sp11:52-53
ammonite from, iridescent in Afghanistan (GNI)F11:238-239; in Opal
(GNI)F11:235-236 Zambia (GNI)Sp11:63-65 from Australia—common
demantoid and topazolite from for sapphire in Sri Lanka (GNI)F11:248- (GNI)W11:319-320; nomenclature
Antetezambato (Pezzotta)Sp11:2-14 249 and characterization


(Beattie)Su11:116 natural, from Tridacna gigas Rhodochrosite
bicolor, bi-pattern hydrophane (Bidwell)Su11:144-145 carving of ancient Indian sage
(GNI)W11:320-322 Pearl simulants (GNI)F11:246-247
blue—(or cristobalite?) from Brazil non-nacreous—(GNI)Sp11:70-71; cal- Rock
(Schnellrath)Su11:142; from Mexico citic “cat’s-eye pearls” fuchsite-rich, deep green (GNI)F11:239-
(GNI)F11:243 (GNI)W11:330-332 240
clarity-enhanced, with artificial matrix Pegmatites Ruby
(LN)W11:312 gem pockets in (Lyckberg)Su11:111-112 geographic origin of (Yetter)Su11:157
from Ethiopia—black (LN)W11:312- Pederneira, in Minas Gerais, Brazil lead-glass-filled—durability and safety
313; dyed purple (Renfro)W11:260- (Pezzotta)Su11:141-142 testing (Sutthirat)Su11:121; trapiche
270; play-of-color (Rondeau)Su11:112 (GNI)Sp11:72-73
Stewart, in Pala, California
sugar-acid treatment of, from Wollo, (Morton)Su11:140 from Liberia (Kiefert)Su11:138
Ethiopia (GNI)W11:333-334
Peridot market for (Drucker)Su11:158-159
Oregon, see United States from pallasitic meteorites—identifica- from Myanmar—color similarity to
Oxygen isotope analysis tion of (Shen)F11:208-213; jewelry Van Gogh’s The Night Café
to characterize gem materials (GNI)F11:243-244 (Dubinsky)Su11:159; overgrowth on
(Rossman)Su11:124-125 from Zabargad, history of painite (Nissinboim)Su11:140-141
(Thoresen)Su11:121-122 from Pakistan, mining of
P variscite from (GNI)F11:251 “pigeon’s blood,” standards for
Painite Phosphorescence rarity of, with regard to treatment and
from Myanmar, with ruby overgrowth in diamond, long-lasting (LN)W11:310- nomenclature (Robertson)Su11:161
(Nissinboim)Su11:140-141 311 from Vietnam, northern (Khoi)F11:182-
Pakistan Photoluminescence spectroscopy, see 195
corundum mining in (GNI)W11:325- Spectroscopy, photoluminescence Russia
326 Plagioclase, see Andesine, Feldspar diamonds from the Nurbinskaya pipe,
sphene from border area with Plastic mineralogical characteristics of
Afghanistan, color-change coating of quartz, imitating emerald (Solodova)Su11:134-135
(GNI)W11:327-328 (GNI)Sp11:68-69
Pallasite, see Peridot Play-of-color, see Opal
Pargasite Postage stamps S
from Tanzania (GNI)Sp11:65 diamond-themed collection St. John’s Island, see Zabargad
Pearl (Overlin)F11:214-219
identification of (Scarratt)Su11:117-119 from Afghanistan, beryllium- and tung-
non-nacreous, from Tridacna gigas sten-bearing (LN)Sp11:53-54
(Bidwell)Su11:144-145 Q from China, blue and pink
Pearl, cultured Quartz (Chen)Su11:136-137
characterization of, using analytical blue, with trolleite and lazulite inclu- blue, with natural Be—(LN)F11:232-233;
techniques (Lu)Su11:149-150 sions (GNI)Sp11:57-58 from Afghanistan (LN)Sp11:53-54
Chinese freshwater—beaded with carving of skull, with inclusions of diffusion-treated with Be—blue, experi-
baroque freshwater cultured pearls izoklakeite (GNI)W11:324-325 ments on (Wathanakul)Su11:153;
(GNI)F11:244-245; with exotic metal- chatoyant—with cinnabar inclusions pink (LN)F11:232; safety and durabil-
lic colors (Beavers)Su11:144; treat- (GNI)W11:322-323; pink, with tour- ity testing of (Sutthirat)Su11:121
ments and their detection maline needle inclusions from Ethiopia (GNI)F11:247-248
(Strack)Su11:120 (GNI)F11:245-246 geographic origin of (Yetter)Su11:157
coated bead-cultured freshwater with emerald inclusions (GNI)W11:323 green, distinct color caused by iron
(LN)W11:313-314 from India, spectral interference in (LN)F11:231-232
grafting methods (Ito)Su11:148 (GNI)Sp11:58-59 irradiation of, electron-beam
identification of (Scarratt)Su11:117-119 plastic-coated, imitating emerald (Seo)Su11:151
marketing of (Overlin)Su11:97-99 (GNI)Sp11:71-72 market for (Drucker)Su11:158-159
from P. margaritifera, absorption fea- spectroscopic differentiation of natural from Pakistan, mining of
tures of (Karampelas)Sp11:31-35 and synthetic (Choudhary)Su11:146- (GNI)W11:325-326
with plastic bead nuclei (LN)F11:229-230 147 “royal blue,” standards for
quality factors of (Ito)Su11:148 with spessartine inclusions (GNI)F11:245 (Atichat)Su11:157-158
sales at the 2010 Myanmar Gem see also Amethyst; Ametrine from Sri Lanka, mining of
Emporium (GNI)Sp11:72-73 [amethyst-citrine]; Quartz, synthetic (GNI)F11:248-249
separation of natural-color saltwater, Quartz, synthetic star, with blue color similar to Van
from P. margaritifera and P. sterna, large, represented as amethyst, aquama- Gogh’s The Starry Night
using spectral differentiation rine, and citrine (GNI)W11:332-333 (Dubinsky)Su11:159
(Karampelas)Su11:117 from Vietnam, northern (Khoi)F11:182-
treatments and their detection 195
Pearl, non-nacreous
R Sapphire, synthetic
green, distinct color caused by cobalt
imitation—calcitic “cat’s-eye pearls” Radioactivity (LN)F11:231-232
(GNI)W11:330-332; skillfully crafted of blue topaz, neutron-irradiated recent advances in growth of (Stone-
(GNI)Sp11:70-71 (Zhang)W11:302-307 Sundberg)Su11:156


Scanning electron microscopy [SEM] (Choudhary)Su11:146-147; amethyst Spodumene
of andesine, reportedly from Tibet (Karampelas)F11:196-201 green cat’s-eye, from Brazil
(Rossman)Sp11:16-30, reflectance, of saltwater cultured pearls (GNI)F11:249-250
(Abduriyim)Su11:167-180 from P. margaritifera Sri Lanka
Scapolite (Karampelas)Sp11:31-35 sapphire from, mining update
from Afghanistan (GNI)Sp11:65-66 of tourmaline from Madagascar, spec- (GNI)F11:248-249
from India (GNI)Sp11:59 tral correlation of manganese and Strain
iron (Ahn)Su11:143-144 in colorless untreated diamonds
Simulants, see specific gem materials
simulated Spectroscopy, laser-induced breakdown (LN)F11:224-225
Social media [LIBS] in HPHT-treated diamonds (Eaton-
and jewelry retailing (Overlin)Su11:95-96 of corundum (Yetter)Su11:157 Magaña)Su11:132
of gem materials (Rossman)Su11:124-125 Sunstone, see Feldspar
automated real-time spectral analysis Spectroscopy, phosphorescence Sustainability
system (Magaña)Su11:154-155 of a diamond with long-lasting phos- Anahí ametrine mine, Bolivia
phorescence (LN)W11:310-311 (Weldon)Su11:163-164
dual integrating sphere, for color grad- Spectroscopy, photoluminescence of coral jewelry (Carmona)Su11:158
ing and measurement (Liu)Su11:154 of cultured pearls, natural-color, P. Symmetry grading
margaritifera and P. sterna assessments and metrics for polished
Spectrometry, laser ablation–inductively (Karampelas)Su11:117
coupled plasma–mass [LA-ICP-MS] diamonds (Caspi)Su11:153-154
of diamond—to establish origin of color GIA boundaries for round brilliant dia-
of andesine, reportedly from Tibet
(Epelboym)Su11:133; HPHT-treated, monds (Geurts)W11:286-295
to identify strain (Eaton-
of aquamarine from Vietnam Synthetics, see specific gem materials
of colored gemstones, and country of Spectroscopy, Raman
origin determination— of aquamarine from Vietnam T
(Abduriyim)Su11:114-116; and appli- (Huong)Sp11:42-48
of garnet (Henderson)Su11:148 Tanzania
cation of discriminant analysis
of odontolite in antique brooches chondrodite from Mahenge
(Krzemnicki)W11:296-301 (GNI)W11:316
of cultured pearls (Lu)Su11:149-150
of pearls—Chinese freshwater cultured garnet from—color-change
of demantoid from northern (GNI)F11:240-241; green grossular
Madagascar (Pezzotta)Sp11:2-14 (Strack)Su11:120; natural and cul-
tured (Scarratt)Su11:117-119 (GNI)Sp11:57-58
of gem materials (Rossman)Su11:124-125 gems from Tunduru
of lavender jadeite from Myanmar, of zircon, brownish red, from China
(Chen)Sp11:36-41 (Limsuwan)Su11:110-111
Guatemala, and Japan pargasite from (GNI)Sp11:65
(Harlow)Su11:116 Spectroscopy, UV-Vis-NIR
see also Tanzanite
of peridot from meteorites, identifica- of andesine (Rossman)Sp11:16-30
tion of (Shen)F11:208-213 of colored gemstones, and country of Tanzanite
of ruby, filled with lead glass origin determination natural-color, identification of
(Sutthirat)Su11:121 (Abduriyim)Su11:114-116 (Smith)Su11:119-120
of sapphire—Be-diffused of diamond, strongly purple-colored Tenebrescence
(Sutthirat)Su11:121; green, distinct black (LN)W11:309 in zircon from Australia (LN)W11:314-
color caused by iron (LN)F11:231-232 of emerald, natural and synthetic (de 315
of synthetic sapphire, green Corte)Su11:147 Thermoluminescence
(LN)F11:231-232 of garnet, color-change, from Tanzania in type IaB diamonds (LN)Sp11:50
of zircon, brownish red, from China (GNI)F11:240-241 Tibet
(Chen)Sp11:36-41 of ruby from Vietnam (Khoi)F11:182- andesine reportedly from (Rossman)
see also Chemical composition 195 [erratum W11:335] Sp11:16-30, (Abduriyim)Su11:167-180
Spectroscopy, electron paramagnetic reso- of tourmaline—(Sriprasert)Su11:152;
Titanite, see Sphene
nance [EPR] from Madagascar, spectral correla-
tion of manganese and iron Topaz
to characterize gem materials
(Ahn)Su11:143-144 color-zoned, yellowish to greenish
of zircon, brownish red, from China brown (GNI)F11:250-251
Spectroscopy, energy-dispersive X-ray fluo- imitation of “mystic” treated topaz
rescence [EDXRF] (Chen)Sp11:36-41
of cultured pearls (Lu)Su11:149-150 Spessartine neutron-irradiated blue, radioactive
Spectroscopy, gamma ray inclusions in quartz (GNI)F11:245 decay pattern of (Zhang)W11:302-307
of topaz, neutron-irradiated blue Sphene [titanite] Topazolite
(Zhang)W11:302-307 color-change, from Pakistan/ from northern Madagascar
Spectroscopy, infrared Afghanistan (GNI)W11:327-328 (Pezzotta)Sp11:2-14
of aquamarine from Vietnam Spinel Tourmaline
(Huong)Sp11:42-48 trapiche, from Mogok, Myanmar from Brazil, Pederneira pegmatite
of diamonds, low-grade, with distinc- (GNI)W11:329-330 (Pezzotta)Su11:141-142
tive color zoning and inclusions from Vietnam—(GNI)Sp11:60-61; inclusions in cat’s-eye quartz
(Kwon)Su11:134 cobalt blue (GNI)W11:328-329 (GNI)F11:245-246
of emerald, natural and synthetic (de Spinel, synthetic from Madagascar, spectral correlation
Corte)Su11:147 star, imitation of moonstone of manganese and iron
of quartz, natural and synthetic— (LN)Sp11:54-55 (Ahn)Su11:143-144


spectroscopic characteristics of United States X-ray diffraction
(Sriprasert)Su11:152 Stewart pegmatite, Pala, California of andesine, reportedly from Tibet
from the United States—California, (Morton)Su11:140 (Abduriyim)Su11:167-180
Stewart pegmatite (Morton)Su11:140; sunstone from Oregon, comparison of
Maine (GNI)Sp11:66-68 two deposits (McClure)Su11:150-151
Trapiche tourmaline from Mt. Marie, Maine— Z
emerald imitation (GNI)Sp11:68-69 (GNI)Sp11:66-68; origin of color
(GNI)Sp11:67-68 Zabargad [St. John’s Island]
ruby, lead-glass-filled (GNI)Sp11:72-73 peridot from, history of
spinel, from Mogok, Myanmar (Thoresen)Su11:121-122
V Zaire, see Congo, Democratic Republic
of the
ancient (Lule)Su11:150 Variscite Zambia
irradiation of blue topaz from Peru (GNI)F11:251 emerald mining at Kagem
(Zhang)W11:302-307 Vietnam (GNI)Sp11:63-65
lead-glass-filling of ruby—and corun- aquamarine from Thanh Hoa
dum triplet (GNI)F11:251-252; Zektzerite
(Huong)Sp11:42-48, (GNI)F11:236-237 gem (GNI)Sp11:61
trapiche (GNI)Sp11:72-73 ruby and sapphire from Tan
pretreatment and heat treatment of Huong–Truc Lau (Khoi)F11:182-195 Zimbabwe
ruby (Lee)Su11:148-149 spinel from—cobalt blue, from Khuoi diamonds from, naturally irradiated—
sugar-acid, of opal from Ethiopia Ngan (GNI)W11:328-329; northern (Crepin)Su11:105; hydrogen-rich
(GNI)W11:333-334 Vietnam (GNI)Sp11:60-61 (Breeding)Su11:129-130
see also Coating; Diamond treatment; Zircon
Diffusion treatment; Dyeing; Filling, brownish red, from China
fracture or cavity; Heat treatment; W (Chen)Sp11:36-41
Irradiation; specific gem materials from Ethiopia (GNI)F11:247-248
Winchell diagrams
Tucson gem shows use in determining garnet composition tenebrescent, from Australia
highlights of (GNI)Sp11:56-63 (Hoover)W11:272-285 (LN)W11:314-315
Turquoise simulants Zoisite, see Tanzanite
odontolite, in set of antique brooches Zoning, see Color zoning; specific gem
(Krzemnicki)W11:296-301 X materials
U of coated bead-cultured freshwater
pearl (LN)W11:313-314
Ultraviolet fluorescence, see Fluorescence, of pearls, natural and cultured
ultraviolet [UV] (Scarratt)Su11:117-119

This index lists, in alphabetical order, the authors of all feature articles, Notes & New Techniques, and Rapid Communications that
appeared in the four issues of Volume 47 of Gems & Gemology, together with the full title and inclusive page numbers of each arti-
cle and the issue (in parentheses). Full citation is given under the first author only, with reference made from coauthors. Also includ-
ed are presenters whose abstracts were published in the Summer 2011 (S11) issue, the proceedings of the Fifth International
Gemological Symposium.

A Ai H., see Chen T. Australian opal nomenclature and

assessment, 116 (Summer)
Amorim H.S., see Schnellrath J.
Abduriyim A.: Geographic origin determi- Anthonis A., see Crepin N. Beavers B., Shepherd J.: Chinese freshwater
nation of colored gemstones, 114-116 Atichat W., Chandayot P., Saejoo S., cultured pearls with exotic metallic col-
(Summer) Leelawatanasook T., Sriprasert B., ors, 144 (Summer)
Abduriyim A., McClure S.F., Rossman Pisutha-Arnond V., Wathanakul P., Benzel W., see Morton D.
G.R., Leelawatanasuk T., Hughes R.W., Sutthirat C.: “Pigeon’s blood” ruby and Berger B., see Morton D.
Laurs B.M., Lu R., Isatelle F., Scarratt K., “royal blue” sapphire: Color standards Bidwell D., DelRe N., Widemann A.,
Dubinsky E.V., Douthit T.R., Emmett for the gem trade, 157-158 (Summer) Epelboym M.: Natural non-nacreous
J.L.: Research on gem feldspar from the Atichat W., see also Huong L.T.-T., pearls from the giant clam Tridacna
Shigatse region of Tibet, 167-180 Wathanakul P. gigas, 144-145 (Summer)
(Summer) Blackman K.N., see Beattie R.A.
Ayalew D., see Rondeau B.
Abduriyim A., see also Choi H.-M. Blodgett T., Gilbertson A., Geurts R.,
Adamo I., see Pezzotta F. Goedert B.: Length-to-width ratios
Ahn Y.-K., Seo J.-G., Park J.-W.: Spectral B among fancy shape diamonds, 129
correlation of manganese and iron in (Summer)
tourmalines from Madagascar, 143-144 Balinskas S., see Magaña Q. Blodgett T., Shen A.: Application of dis-
(Summer) Barjon J., see Willems B. criminant analysis in gemology:
Ahn Y.-K., see also Kwon S.-R., Seo J.G. Beattie R.A., Blackman K.N., Levonis H.: Country-of-origin separation in colored


stones and distinguishing HPHT-treated erties of silicon-related defects in CVD tion of high-value colored stones, 108-
diamonds, 145 (Summer) synthetic diamond, 131-132 (Summer) 110 (Summer)
Blodgett T., see also Geurts R.H. Danisi R.M., see Pezzotta F. Goedert B., see Blodgett T., Shigley J.E.
Bodeur Y., see Rondeau B. DelRe N., see Bidwell D., Epelboym M. Grambole D., see Nasdala L.
Breeding C.M.: Hydrogen-rich diamonds Diella V., see Pezzotta F. Green B., see D’Haenens-Johansson U.F.S.
from Zimbabwe with natural radiation Dobrinets I., see Epelboym M. Grizenko A.G., see Vins V.G.
features, 129-130 (Summer) Donaldson C.H., see Taylor R.P. Groover K., see Morton D.
de Brum T.M.M., see Juchem P.L., Douman M., see Kiefert L. Gumpesberger S.M.: Extreme conoscopy,
Schnellrath J. Douthit T.R., see Abduriyim A. 147 (Summer)
Buckley S., see McManus C. Dowe J., see McManus C.
Downs R.T., see Henderson R.R.
Drucker R.B.: Market trends in a changing
C global economy, 158-159 (Summer) Häger T., see Huong L.T.-T.
Carmona C., Cole J.E., Marks J.: “Too pre- Dubinsky E., King J., Zucker B.: Gemstone Hainschwang T., see Fritsch E.,
cious to wear”: The role of the jewelry color as nature’s palette: Van Gogh’s com- Karampelas S.
industry in coral conservation, 158 mentary on a 111 ct star sapphire and Harlow G.E., Shi G.-H.: An LA-ICP-MS
(Summer) other gem-quality corundum, 159 study of lavender jadeite from Myanmar,
Caspi A.: Inclusion mapping in diamonds, (Summer) Guatemala, and Japan, 116-117 (Summer)
153 (Summer) Dubinsky E., see also Abduriyim A. Harlow G.E., see also Nissinboim A.
Caspi A., Kerner A.: Symmetry: From Harrell J.A., see Thoresen L.
assessments to metrics, 153-154
Harris J.W., see Nasdala L.
(Summer) E Henderson R.R., Downs R.T., La Sure J.:
Chandayot P., see Atichat W.
Eaton-Magaña S.C.: Observation of strain Determining the chemical composition
Chapman J.: Recent studies of colored dia- through photoluminescence peaks in of garnets using Raman spectroscopy,
monds from Argyle, 130 (Summer) diamonds, 132 (Summer) 148 (Summer)
Chen H., see Zhang B., Zhang J. Edmonds A.M., see D’Haenens-Johansson Herzog F., see Krzemnicki M.S.
Chen T., Ai H., Yang M., Zheng S., Liu Y.: U.F.S. Hofmeister W., see Huong L.T.-T., Nasdala L.
Brownish red zircon from Muling, Eickhorst M.: The impact of the choice of Hoover D.B.: Determining garnet composi-
China, 36-41 (Spring) illumination in the wholesale buying, tion from magnetic susceptibility and
Chen T., Yang M., Ai H.: Blue and pink selling, and grading of gemstones, 123 other properties, 272-285 (Winter)
sapphires from Muling, northeastern (Summer) Hughes R.W., see Abduriyim A.
China, 136-137 (Summer) Emmett J.L., see Abduriyim A. Huong L.T.-T., Hofmeister W., Häger T.,
Chodur N.L., see Juchem P.L. Epelboym M., DelRe N., Widemann A., Khoi N.N., Nhung N.T., Atichat W.,
Choi H.-M., Kim Y.-C., Kim S.-K., Zaitsev A., Dobrinets I.: Characterization Pisutha-Arnond V.: Aquamarine from
Abduriyim A.: Evidence of an interstitial of some natural and treated colorless and the Thuong Xuan District, Thanh Hoa
3H-related optical center at 540.7 nm in colored diamonds, 133 (Summer) Province, Vietnam, 42-48 (Spring)
natural diamond, 131 (Summer) Epelboym M., see also Bidwell D.
Choi K.-M., see Lee B.-H.
Choudhary G.: Remarkable gems encoun- I
tered at the Gem Testing Laboratory, F
Jaipur, India, 145-146 (Summer) Isatelle F., see Abduriyim A.
Choudhary G., Fernandes S.: Spectroscopic Fallick A.E., see Giuliani G. Ito M.: Circle and spot formation mecha-
examination of commercially available Feneyrol J., see Giuliani G. nisms and changes in luster, color, and
quartz varieties: A gemological perspec- roundness of cultured pearls by grafting
Fernandes S., see Choudhary G.
tive, 146-147 (Summer) methods in Pinctada margaritifera, 148
Finch A.A., see Taylor R.P. (Summer)
Clanin J.C.: The fundamentals in mining Fleck R., see Morton D.
for colored gemstones and mineral speci- Iverson J.:
Fritsch E., Massuyeau F., Rondeau B., Great expectations, 259 (Winter)
mens, 107-108 (Summer) Segura O., Hainschwang T.: Advancing
Cole J.E., see Carmona C. New beginnings, 181 (Fall)
the understanding of luminescence in
Coopersmith H.G.: The making of a dia- gem materials, 123 (Summer)
mond mine: Why everyone cannot have Fritsch E., see also Karampelas S., Rondeau B.
one, 108 (Summer) J
de Corte K., Van Meerbeeck M.: IR and Janse B.: Mystery diamonds—From alluvial
UV-Vis spectroscopy of gem emeralds— G deposits, 110 (Summer)
A tool to differentiate natural and syn- Jia Q.: An analysis of the development and
thetic stones, as well as geographic ori- Gasanov M., see Solodova Y. tax policy of China’s gem industry, 159-
gin? 147 (Summer) Gatta G.D., see Pezzotta F. 160 (Summer)
Crepin N., Anthonis A., Willems B.: A case Gauthier J.-P., see Karampelas S. Johnson M.: Mineral collectors—An over-
study of naturally irradiated diamonds Geurts R.H., Reinitz I.M., Blodgett T., looked gem market, 160 (Summer)
from Zimbabwe, 105 (Summer) Gilbertson A.M.: GIA’s symmetry grad- Johnson P.: Pink HPHT synthetic dia-
ing boundaries for round brilliant cut monds—A new coloration technique,
diamonds, 286-295 (Winter) 133 (Summer)
D Geurts R., see also Blodgett T. Juchem P.L., de Brum T.M.M., Chodur
D’Haenens-Johansson U.F.S., Edmonds Gilbertson A.M., see Blodgett T., Geurts R.H. N.L., Liccardo A.: Gem materials in the
A.M., Green B., Newton M.E., Mar- Giuliani G., Ohnenstetter D., Fallick A.E., south of Brazil, 137-138 (Summer)
tineau P.M., Khan R.U.A.: Optical prop- Feneyrol J.: State of the art in the forma- Juchem P.L., see also Schnellrath J.


K Levonis H., see Beattie R.A. Kimbrough D., Berger B., Fleck R., Kistler
R., Benzel W., Snee L., Groover K.:
Lhuaamporn T., see Wathanakul P.
Karampelas S., Fritsch E., Gauthier J.-P., Liccardo A., see Juchem P.L. Petrogenesis of the Li-bearing Stewart
Hainschwang T.: UV-Vis-NIR Likes T., see McManus C., Yetter K. pegmatite, Pala, California, 140 (Summer)
reflectance spectroscopy of natural-color Mosselmans F.J., see Taylor R.P.
Limsuwan R., Sakimoto T.: Geology,
saltwater cultured pearls from Pinctada
exploration, mining, and processing of
margaritifera, 31-35 (Spring)
gems and gold from placer deposits on
Karampelas S., Fritsch E., Hainschwang T., the Muhuwesi River, Tunduru, N
Gauthier J.-P.: Spectral differentiation of Tanzania, 110-111 (Summer)
natural-color saltwater cultured pearls Nam N.V., see Khoi N.N.
Limsuwan R., Sakimoto T.: Using a con- Nasdala L., Grambole D., Harris J.W.,
from Pinctada margaritifera and Pteria
crete mixer to sort gem-bearing gravels Schulze D.J., Hofmeister W.: Radio-col-
sterna, 117 (Summer)
for artisanal and small-scale mining, 139 oration of diamond, 105-106 (Summer)
Karampelas S., Fritsch E., Zorba T., (Summer)
Paraskevopoulos K.M.: Infrared spec- Newton M.E.: Treated diamond identifica-
Lithgow G., see McManus C. tion, 106 (Summer)
troscopy of natural vs. synthetic
amethyst: An update, 196-201 (Fall) Liu Yan: Instrumental color measurement Newton M.E., see also D’Haenens-
and grading of faceted gemstones, 154 Johansson U.F.S.
Ke J., see Zhang B.
(Summer) Nhung N.T., see Huong L.T.-T., Khoi N.N.
Keller A., see Kimmel K.
Liu Yungui, see Chen T. Niklewicz D., Whetstone W., Matula L.:
Kerner A., see Caspi A.
Lu R., Zhou C.-H., Sturman N.: Hallmarking: A powerful benefit to the
Khan R.U.A., see D’Haenens-Johansson Operational considerations of EDXRF,
U.F.S. consumer, retailer, and international
LA-ICP-MS, and photoluminescence trade, 160-161 (Summer)
Khoi N.N., Sutthirat C., Tuan D.A., Nam techniques in the analysis of pearls, 149-
N.V., Thuyet N.T.M., Nhung N.T.: Nissinboim A., Harlow G.E.: A study of
150 (Summer) ruby on painite from the Mogok Stone
Ruby and sapphire from the Tan Lu R., see also Abduriyim A.
Huong–Truc Lau Area, Yen Bai Tract, 140-141 (Summer)
Lu T., see Zhang B., Zhang J.
Province, northern Vietnam, 182-195
(Fall) Luke A.:
Khoi N.N., see also Huong L.T.-T. Everything old is new again: The appeal O
of the auction, estate, and vintage
Kiefert L., Douman M.: Ruby from Liberia, Ohnenstetter D., see Giuliani G.
markets, 100-102 (Summer)
138 (Summer) Ostrooumov M.: Gemstones from
Jewelry design: From the masses to
Kim K.Y., see Seo J.G. Mexico—A review, 141 (Summer)
museums, 92-94 (Summer)
Kim S.-K., see Choi H.-M. Ottaway T.L., see Shigley J.E.
Playing a bigger game: Better business
Kim Y.-C., see Choi H.-M., Lee B.-H. for a better world, 103-104 (Summer) Overlin S.D.:
Kimbrough D., see Morton D. Lule C.: Experiments on ancient gem treat- Digital age marketing: Are you still par-
Kimmel K., Keller A.: GIA Symposium ment techniques, 150 (Summer) tying like it’s 1999?, 95-96 (Summer)
2011: Advancing the science and busi- Lyckberg P.: Gem pockets in pegmatites— The future of gemstones and gemstones
ness of gems, 79 (Summer) A worldwide study and comparison, of the future, 97-99 (Summer)
King J., see Dubinsky E. 111-112 (Summer) A history of diamonds through philately:
Kistler R., see Morton D. The Frank Friedman collection, 214-
Klettke J.A.: Systematic diamond descrip- 219 (Fall)
tions for use in geology and exploration, M Navigating the current economy for busi-
139 (Summer) ness growth and success, 80-82
Magaña Q., Balinskas S.: Automated real- (Summer)
Koivula J.I., see Shen A. time spectral analysis for gemstones,
Kozlov A.V., see Vasilyev E. Shipwrecked! 126-128 (Summer)
154-155 (Summer)
Krzemnicki M.S., Herzog F., Zhou W.: A Where is luxury in this brave new
Marks J., see Carmona C.
historic turquoise jewelry set containing world? 89-91 (Summer)
Martineau P.M., see D’Haenens-Johansson
fossilized dentine (odontolite) and glass, U.F.S.
296-301 (Winter)
Krzemnicki M.S., see also Strack E.
Massuyeau F., see Fritsch E. P
Matula L., see Niklewicz D.
Kurbatov K., see Solodova Y. Paraskevopoulos K.M., see Karampelas S.
Mazzero F., see Rondeau B.
Kwon S.-R., Seo J.-G., Ahn Y.-K., Park J.- Park J.-W., see Ahn Y.-K., Kwon S.-R., Seo J.G.
McClure S.F.: A gemological comparison
W.: Characterization of distinctive color Petrovsky V.A., see Vasilyev E.
of the two major Oregon sunstone
zoning and various inclusions in low- Pezzotta F., Adamo I., Diella V.:
deposits, 150-151 (Summer)
grade diamonds, 134 (Summer) Demantoid and topazolite from
McClure S.F., see also Abduriyim A.,
Renfro N. Antetezambato, northern Madagascar:
Review and new data, 2-14 (Spring)
L McManus C., Likes T., McMillan N.,
Yetter K., Dowe J., Wise M., Buckley S., Pezzotta F., Adamo I., Diella V., Gatta
La Sure J., see Henderson R.R. Lithgow G., Stipes C., Torrione P.: G.D., Danisi R.M.: The Pederneira peg-
Laurs B.M., see Abduriyim A. Determining the provenance of gem- matite, Minas Gerais, Brazil: Geology
Lee B.-H., Choi K.-M., Kim Y.C., Yon S.-J.: stones using the Materialytics and gem tourmaline, 141-142 (Summer)
The observation of defects after the pre- Sequencing System (M2S), 155-156 Pisutha-Arnond V., see Atichat W., Huong
treatment and simple heat treatment of (Summer) L.T.-T., Wathanakul P.
ruby, 148-149 (Summer) McManus C., see also Yetter K.
Lee C.-T., see Morton D. McMillan N., see McManus C., Yetter K.
Leelawatanasuk T., see Abduriyim A., Miceli R.S.D., see Schnellrath J.
Atichat W., Wathanakul P. Morton D., Sheppard B., Lee C.-T., Reinitz I.M., see Geurts R.H.


Renfro N., McClure S.F.: Dyed purple Sriprasert B.: Spectroscopic characteristics Experiments on heating of blue sap-
hydrophane opal, 260-270 (Winter) of some gem tourmalines, 152 phires with beryllium, 153 (Summer)
Robertson S.: A case study in the percep- (Summer) Wathanakul P., see also Atichat W.
tion of rarity—Ruby, 161 (Summer) Sriprasert B., see also Atichat W., Weldon R.: The Anahí ametrine mine: A
Rondeau B., Fritsch E., Mazzero F., Ayalew Wathanakul P. model for gemstone sustainability, 163-
D., Bodeur Y.: Amazing play-of-color Stachel T.: Diamonds and cratons—Does 164 (Summer)
opals from Wollo, Ethiopia—Toward a the relationship hold for the Canadian Whetstone W., see Niklewicz D.
genetic model, 112 (Summer) deposits? 112-114 (Summer) Widemann A., see Bidwell D., Epelboym M.
Rondeau B., see also Fritsch E. Steinbach M.P.: New artificially asteriated Willems B., Tallaire A., Barjon J.:
Root C.K.: Abbreviated jewelry descrip- gemstones, 152-153 (Summer) Exploring the origin and nature of lumi-
tions—A uniform standard, 161-162 Stipes C., see McManus C. nescent regions in CVD synthetic dia-
(Summer) Stone-Sundberg J.: Recent advances in syn- mond, 202-207 (Fall)
Rossman G.R.: thetic sapphire growth, 156 (Summer) Willems B., see also Crepin N.
Advanced gem characterization, 124-125 Strack E., Krzemnicki M.S.: Experimental Wise M., see McManus C.
(Summer) treatments involving dye and heat for Wongkokua W., see Wathanakul P.
The Chinese red feldspar controversy: Chinese freshwater cultured pearls, and
Chronology of research through July their detection with Raman spectroscopy,
2009, 16-30 (Spring) 120 (Summer) Y
Rossman G.R., see also Abduriyim A. Sturman N., see Lu R.
Sukharev A., see Vasilyev E. Yager T.R.: Recent trends in world colored
gem production, 142-143 (Summer)
Sutthirat C.: Durability and safety testing
S of lead glass–treated ruby and beryllium- Yang M., see Chen T.
treated sapphire, and environmental Yelisseyev A.P., see Vins V.G.
Saejoo S., see Atichat W. Yetter K., McMillan N.J., McManus C.,
impact assessments, 121 (Summer)
Sakimoto T., see Limsuwan R. Likes T.: Provenance of rubies and sap-
Sutthirat C., see also Atichat W., Khoi N.N.
Samosorov G., see Solodova Y. phires: an application of laser-induced
Scarratt K.: Pearl identification—A practi- breakdown spectroscopy (LIBS) and
tioner’s perspective, 117-119 (Summer) T advanced chemometrics for the gem
industry, 157 (Summer)
Scarratt K., see also Abduriyim A.
Schnellrath J., Amorim H.S., Juchem P.L., Tallaire A., see Willems B. Yetter K., see also McManus C.
de Brum T.M.M., Miceli R.S.D.: Blue Tashey T.E.: Recommendations for advanc- Yon S.J., see Lee B.-H.
“opal” (or cristobalite?) from Rio Grande ing the color science of gemology, 163
do Sul, Brazil, 142 (Summer) (Summer)
Schulze D.J., see Nasdala L. Tawornmongkolkij M., see Wathanakul P. Z
Sedova E., see Solodova Y. Taylor R.P., Finch A.A., Donaldson C.H.,
Mosselmans F.J.: Applications of fluores- Zaitsev A., see Epelboym M.
Segura O., see Fritsch E. Zhang B., Lu T., Chen H., Ke J.: Research
cence in gemology, 125 (Summer)
Seo J.-G., Kim K.Y., Ahn Y.K., Park J.-W.: and identification of natural and treated
Changes in color and optical properties Thoresen L., Harrell J.A.: Archaeo-
gemology of peridot, 121-122 (Summer) nephrite in China, 122 (Summer)
of various sapphires by electron-beam Zhang J., Lu T., Wang M., Chen H.: The
irradiation, 151 (Summer) Thuyet N.T.M., see Khoi N.N.
radioactive decay pattern of blue topaz
Seo J.-G., see also Ahn Y.-K., Kwon S.-R. Torrione P., see McManus C.
treated by neutron irradiation, 302-307
Serras-Herman H., Herman A.: Lectures— Tuan D.A., see Khoi N.N. (Winter)
An inspirational way for artists to com- Zheng S., see Chen T.
municate, 162 (Summer) Zhou C.-H., see Lu R.
Shen A., Koivula J.I., Shigley J.E.: V Zhou W., see Krzemnicki M.S.
Identification of extraterrestrial peridot Van Meerbeeck M., see de Corte K. Zorba T., see Karampelas S.
by trace elements, 208-213 (Fall) Vasilyev E., Petrovsky V.A., Kozlov A.V., Zucker B., see Dubinsky E.
Shen A., see also Blodgett T. Sukharev A.: The nature of black col-
Shepherd J., see Beavers B. oration in gem-quality diamonds from
Sheppard B., see Morton D. Brazil, 135 (Summer)
Shi G.-H., see Harlow G.E. Vins V.G.: New ultra-deep diamond clean-
Shigley J.E., Goedert B., Ottaway T.L.: The ing technology, 136 (Summer)
GIA Gem Project—An online resource Vins V.G., Yelisseyev A.P., Grizenko A.G.:
of information on gemstones, 151-152 Donor nitrogen aggregation in synthetic
(Summer) diamonds annealed in the graphite sta-
Shigley J.E., see also Shen A. bility field, 135-136 (Summer)
Shor R.: Polished diamond supplies from
secondary market sources, 163
(Summer) W
Smith C.P.: Natural-color tanzanite, 119- Wang M., see Zhang J.
120 (Summer) Wang W.: A review of diamond color treat-
Snee L., see Morton D. ment and its identification, 107
Solodova Y., Sedova E., Gasanov M., Samo- (Summer)
sorov G., Kurbatov K.: Mineralogical char- Wathanakul P., Wongkokua W., Lhuaam-
acteristics of diamonds from the Nurbin- porn T., Tawornmongkolkij M., Leela-
skaya pipe, Yakutian diamond-bearing watanasuk T., Atichat W., Pisutha-
province, Russia, 134-135 (Summer) Arnond V., Sutthirat C., Sriprasert B.: